Graduated and proportional symbol maps are two variations on the same idea, which is essentially that you're going to have a symbol where the size of that symbol represents its value. So, the bigger the symbol, the higher the value, the lower the value, the smaller the symbol, you get the idea. So, let's have a look at how that works. So, if I right-click on my feature class and go to Properties and the Symbology tab, you'll see here with graduated colors, also known as a choropleth, that I've been using the population density value for my map. If I switch to graduated symbols, I can still use that same population density attribute and I have the same classification methods available to me, so I can use natural breaks, I can use equal interval, quantile, and so on. I'm just going to stick with natural breaks and five classes for now. The main options we have available with graduated symbols is that we can change the symbol sizes. So, the smallest symbol will be four points, that's that one there. The largest symbol will be 18 points, which is that one there. Then it automatically calculates sizes in between to go with that, and we can change the color of the symbol, and we can change the color of the background. So, I'll leave the symbol sizes the way they are for now, I'm just going to change the color. Let's pick something that's similar to what I'm already using, let's say it's purple, and say Okay. Then the background, I'm going to change to be a light gray. The idea is that I want the purple symbols to really stand out against the background. Say Okay, and so there we go. So actually, the background didn't work out so great because now it looks the same as the rest of the background of the map, so I'm not getting a lot of figure-ground contrast there. I could adjust that, but the idea is that we are certainly seeing that the higher population density values that we would expect to see in the downtown area have larger symbols, the areas with smaller values or lower population density, as we move towards the suburbs, have smaller symbols. So, this is a nice visual way of being able to show that pattern in a way that's different than typical choropleth. Of course, you can use graduated or proportional symbols for points. I happen to be using them for areas here just because it's something different. So, this is Graduated symbols, let's go back and try proportional symbols. The difference is that remember with graduated symbols, we're actually classifying the data into different classes. With proportional symbols, we're not. What we're going to do is use our same density values, but now every single value will have its own sighs. If we have 50 different values, there will be 50 different sizes of symbols. Every single one of them will be scaled to match the size of the value in that dataset. So, all we're really able to do here is set the size of the minimum value, and then it will show you what the maximum value will look like on the map. So, this is going to be much larger symbol than the graduated symbol one. It's probably going to look like a bit of a mess. But there's this balance has to go on there in terms of the range of the data that you have, is that if I make the minimum symbol size too small, you won't be able to see a lot of the smaller data values, and so we end up with maximum value that's a little too big. But let's just see how this looks. I'm going to change the color to my purple again, say Okay, I'll change the background to, let's try a beige color here, and a matching beige outline, and say Okay. So, this is a very different-looking map in the graduated symbol one. In some ways, I almost like it because you get to see this visual clustering or density of values in the downtown area, obviously I think it's a little too much. But some things to notice here is that you do have these values in the legend or the symbols in the legend which are just representative symbol sizes. They're just telling you that if you have a value of 100,000, that's how big the symbol would be. If you had a value of 1,000, that's how big the symbol would be, and so on. Lets go back and see if we can adjust this a little bit. So, we could go down to an even smaller symbol size. I don't think this is going to look great, but let's just have a look at what this will look like, say Okay. Actually, it almost works. So, the smaller symbols are still visible, so you can see those smaller symbols and now we are getting clustering downtown, but it's not too bad. So, we're getting that gradation of different sizes for different areas for different values, so this would be a viable alternative to the graduated symbol. So, the main trade-off here is that graduated symbol maps are a little easier to interpret especially if you have say three or four or five classes, it's fairly easy for people to visually or mentally slot those into different things like small symbol, small value, medium symbol, medium value. With proportional symbol, you're getting a little more nuanced version of the data, you're actually able to see all of the different sizes, and so that can be a good thing in terms of you getting more gradations of values and sizes, but sometimes that can be a little harder for people to interpret. There's not necessarily one right way to do that, you can experiment with both of them and play with your data. But with any of these map mapping methods, you really have to look at the statistical distribution, as well as the geographic distribution of your data, and experiment with these different data types, different classification methods, different sizes, even different colors, to try and find one that's going to work well for your particular application.
