A very important concept related to visual channels is the idea that not all channels are equally effective at representing information. In particular, in visualization we like to talk about two broad classes of properties. One is expressiveness and the other is effectiveness. So, expressiveness is about what can be expressed with a channel, and effectiveness is about how well the channel can express this information. So, let's focus first on expressiveness. So, expressiveness is the type of information that a given channel can or cannot express. So, before I give you examples and explain these aspect in more details, I need to clarify what I mean when I say type of information. What do we mean by type of information? So, let's do that first. Well, in statistics, there are different ways of describing informational data. And typically, there are four classes that are used. So, the idea is that information or a measurement can be either nominal, ordinal, interval, or ratio. So, what does it mean? So, a nominal measurement means that basically, these information is about a number of categories. The only operation that we can do with categories is comparing them to see whether a category is the same as another category. Then we have ordinal information. What is ordinal information? Well, ordinal is very similar to categorical, to nominal, but the difference is that we can identify an order in the elements, but we can't really tell anything about how much bigger or how much lower a given element is compared to another. So, any ranking is an example of ordinal measurement or data, or for instance you see when measurements are extracted from surveys, when we ask people to answer a question about something and they have to specify say, medium, high, low something, well, that would be an example of a of an ordinal scale. Because why is it an ordinal scale? Well, because we can't really tell what's the quantitative difference between these elements but we know that there is an order. Then we have interval and ratio scales. So, interval and ratio scales are about measuring quantities, both about measuring quantities. But the main difference is that interval scales don't have a zero value, and because of that there are certain operations don't make sense, like for instance, division doesn't make sense. What are examples of interval measurements? Well, for instance percent or temperature are examples of interval measurements. Then we have racial. So, for instance measurements like weight or height of a person is an example of a ratio scale. So, in visualization, we like to categorize these elements, these type of measurements in a smaller set. Let me tell you what the set is first and then why it is useful in visualization. Well, the set is quantitative information, sequential information, and categorical information. So, why do we do that? Well, because when someone has to decide what is the appropriate visual representation or the appropriate visual channel to represent a given piece of information, it's very important to know the properties of this information. And in particular, whether this information is about a quantity, it's about a sequence, or about categories, and that's the reason why we talk about quantitative data, sequential data, and categorical data. So, let's go back to the concept of expressiveness. So, let me say again what expressiveness is. So, expressiveness is about the type of information that can or can't be expressed by a channel. And more precisely now after explaining what are different types of measurements or different types of data, we can say that expressivenes is about whether a channel can express information about quantities, sequences or categories. At the same time whether a given channel while expressing some of this information is also not expressing something that is unintended. Something that I'm going to explain later on in more details. Before I conclude, I want to give you a couple of examples. So, here you see that I have four different colors. Imagine that I want to encode information with the channel color. More precisely, color U, which is the name of the color. So, now, what kind of information can I express with color? Can I express quantity? Not really. Because these colors don't really have an order. You can't really say that green is bigger than red for instance. You don't perceive visually this information as order or quantitative. So, you can't really express sequences or order and you can't really express magnitude or quantities with this colors or with color in general, with color U. So, that's one example of the fact that what is the expressiveness of color U? Color U can't really express quantity and sequence but it can express the idea of categories. It can express nominal data, nominal information. Here is another example. The length of bars. This is something that we used in the previous examples. Okay, so length of bar communicates the idea of quantity. So, the bigger the bar, the bigger the value. It can also represent the order. So, in this case, I put these bars in order on purpose so that they convey the idea of order. So, they can be instrumental in representing the idea that something is ordered. Can the length of a bar represents categories? Not really. One may try to do it would not really represent this information in a way that is perceived, that is naturally perceived by a human observer. So, these are two small examples that explain the idea of expressiveness. Once again, the expressiveness concept is about whether a channel can express certain types of information or not.
