PL-300 – Section 4: Part 1 Level 3: Creating different visualizations: Matrices and bar charts Part 2

32. Bar Chart Formatting, Including Continuous Versus Categorical Axes

Just a quick look at all of the formatting that you can do. So, you can change the X-axis, so you can have either Continuous or a Categorical type. So, you can see the difference. What this means is categories each year becomes a unique category as opposed to something on a scale. So, here we treating it something on a scale from 1995 through to 2015. Here it’s treating it like a category just like red, green, blue, or car, vehicle, lorry. Each of those are categories.

If it was continuous, then you get to set the start and end values. And, if you have got logarithmic data, so data that goes 1, 10, 100, 1,000, which really multiplies rather than just in 10s, rather than just adds like ours does in ones, then you can choose Logarithmic but if you don’t have that data, just keep to Linear. So, we’ve got that for the X-axis.

We’ve got similar things for the Y-axis. You’ll see that Linear is the only one that the computer thinks makes any sense, but you can change the Start and the End values as well. And you can have the position of where the axis is. So here it is on the right-hand side. Here it is on the left. Now, scrolling down, you can change other things like the Display units.

Probably quite useful. You can see it’s automatically displayed in thousands. You may say, “No, I want it to display as the full figure,” or “I want it to display in millions.” So, you can have an axis Title. So, here we have SalesVolume on and off, and you can say, “Show the title only,” or “Show the unit only.” Here you can see the word thousands, or show both.

You can edit the axis Title if you’ve got one. You can edit the Size, add Gridlines or remove Gridlines. The gridlines are these little lines going across and how thick the gridlines are on a scale of one to four and whether they are solid, dotted, or dashed.

Looking at some of the other things, so we’ve got a legend. So, whether we see the Legend that’s this up here and where it is positioned. So you can have it in quite a few places. Whether you’ve got the Title for the legend, in this case, Region Name, or whether that’s not needed and, of course, we can adjust the Title.

We’ve had a look at adjusting the Data colours previously.

You can have Data labels. Now, Data labels allow you to see instantly what that particular value is. So, it gets written, in this case, inside the data label, right in the middle, but you could have it sent right at the top, or sent to middle, or sent to bottom. So, it depends where you want it.

Now, you may notice some of the labels aren’t showing. So, the reason for this is because it thinks, “Okay if I show too much, that’s going to, perhaps, annoy the end user.” So, you could change the Label density so that it only shows certain labels, or change it to 100% so it shows all. So, if you’re struggling and going why aren’t all of them being shown, have a look at the Label density.

There’s also a Background. So, if you want the label to have its own background like this, you can also do this as well.

Just going down, we’ve also got the Plot area background. So, here we can add an image to the background, if you wish, and change the transparency of the background. So, if you had, I don’t know, a sunset or your corporate logo that could be in the background washed as much as you wish.

Now, in addition to all of those, which I have just gone through, there are also a lot of those particular categories that we’ve had a look at previously. I won’t be pointing them out each time. So, there will be a Border, there will be Tooltips you can add, there will be Titles, there will be General where you can actually fine tune it. Here you can see it’s not quite all the way to the left.

So, this is a look at all of the formatting that you can do for Stacked column charts and Stacked bar charts.

33. Configure Interactions Between Visual (Edit Interactions)

Now, in this video, I want to overlook at the interactivity between two different visualisations. So, what I’m going to do is I’m going to copy this visualisation, and go back to our original visualisation, the table. I’m just going to get rid of this second one, which we’ve now included in the matrix, and paste it, resize. And you could also do lots of other things as you wished as well.

Now, we have got here one table visualisation, and one bar chart visualisation. Now, if I click on Greater Manchester in the table, notice what happens in the bar visualisation. And it’s actually a bit difficult to see what does indeed happen, but you can see something definitely has, if I deselect.

Now, the reason it’s a bit difficult, is because all of these colours are sort of greyed out anyway. So, if I click on the formatting into data colours, and just reset them, revert to default, you’ll see that we’ve got vibrant colours to begin with. And then when I select Greater Manchester, all of the other colours get washed out apart from Greater Manchester. Going to Merseyside, and you can see they get highlighted.

Now, we can edit what happens between visualisations. If I click on one of the visualisations, the one that I’m going to be selecting the various components, and I go to Visual Tools, Format, Edit interactions. So, you see what it says, change how visualisations interact when data points are selected. So, I’ll click on this Edit interactions. And what happens is it gives me an extra three icons. And you’ll see they expand as soon as I get onto them, but it really is these three icons, I want to talk about. We’ve got Highlight. Highlight, in this case, is the default. So, if I click on a particular region name, we are looking at that region.

But the other one, other main one I wanted to talk about, is Filter. If I click on this, nothing appears to happen. But now, I’m going to click on South Yorkshire, and the bar changes. The bar chart. So that it is solely about South Yorkshire. Click on Merseyside, and it’s solely about Merseyside.

Now, the third interaction is None. So, that means I can click on any of these, and it doesn’t actually affect this lower visualisation. So, you can select what happens when a particular visualisation gets clicked on, what actually happens. Now, this is very useful when you have additional visualisations on a particular page, more than one.

So, I’m going to insert a new table visualisation. So, I’ll just click on table. There it is. And I’m going to add in the date, just the year, and I’m going to have a total of the sales volume. So, let’s just format it so that it’s a lot bigger. So, what I can do now is click on South Yorkshire, and that 3,795,000 total gets reduced to 431. Click on Merseyside, and that gets reduced as well. So, what happens if I’ve got South Yorkshire, and I click on 2000? So, you’ll see that only one selection happens at once. So, I can’t click on time and where, and 2002. As soon as I click on something else, the other one gets deselected. Well, I can, if I use the Ctrl, but more about that later.

Now, another thing to bear in mind, is that if you’ve got visualisations which are very close together, then you might not see the Highlight and Photo icons. To remedy this, let’s just move this table up. And you can see that they are hiding behind this visualisation. So, if you are doing this Edit interactions, then you might just temporarily want to have a bit of vertical space between the visualisations, just to make sure that you can see all these icons okay. So, as I’ve still got these Edit interactions, I can say, okay, with this, I want you to do a Filter.

So, here we have a Filter. So, now if I select 2003, it focuses solely on 2003, but you can also see that we can also interact with other tables. So, here we are filtering this second table, or this first table, through the second table. So, when I click on a particular year, it just shows what’s in that year. So how you can use this combination of highlighting formatting is really only defined by your ambition for this particular page. What do you want it to do when you click on a particular number? Do you want it to filter? Or do you want it to highlight? So, I’ll change this one back to a highlight. And so you can see it highlights year 2000. So, this could be quite a useful demonstration tool. So, you could say, well, let’s just focus on 2006. Here was the result.

Now, let’s add 2007, I’m going to hold down Ctrl, and select 2007, and 2009. And so you can select more than one thing from a particular visualisation. Now, if you do want to have, say South Yorkshire in 2006, ‘07 and ‘09, then you can do that using the Ctrl, like this. So, here I’ve got South Yorkshire of 2006, ‘07, ‘09, all highlighted. Now, I’ve added West Midlands to the mix, and I’m going to deselect West Midlands. All of that through using the Ctrl. So, this makes it quite a powerful tool to be able to look at individual items, or more than one item at once, and just focus down on this.

Now, you may have noticed in off applications, I could click on the first one, and then hold down Shift, and click on the last to get a range. That is not a possibility here. You’ve got use the Ctrl to select multiple items.

So, this Edit interactions, once you’ve finished, deselected, the interactions will still happen, but you’ll no longer have those little icons that enable you to change what they do from Highlight to Photo, or to None.

So, interactions, very useful on one particular page. They allow you to focus on a particular topic, and you can narrow it down by using additional tables to focus the end user’s attention. And just in case you are thinking, can we do it either way? Yes. If I click on a particular area of the graph, we can change the tables so that it’s focusing on Tyne and Wear for the year of 2004, for example. And then holding down Ctrl, I can add additional years, or additional other areas as well. So, all of that can be tailored with the Edit interactions.

34. Clustered and 100% Stacked Bar Charts

Now, in the previous videos, we’ve had a look at Stacked bar and Column charts. And the essence of these is that the individual elements get put on top of each other. So, this action, which is 17,000, gets put on top of another one, which is 33,000. And together they get the 50,000 line.

But, suppose, you wanted each to be on an individual space. We can do that with the Clustered column charts. So, if I just select this particular item, and I convert it to a clustered column chart. Obviously, you can also insert one, as well, if you so wish. Here you can see the result. And you can see it there’s an awful lot of information. So, I don’t recommend using Clustered column charts where there is this much information.

If, however, I was to Drill down a bit so that instead of looking at all of the data, we would just be going down to the next level. Here we have the quarters and even the months. That is a fairly reasonable amount of data that can be shown. And here we can see, for instance, the seasonality of Greater Manchester. And see wherever there was a similar seasonality of Tyne and Wear. Whereas previously, if I just click undo a few times, it was difficult to see the seasonality of Tyne and Wear because the places were, the individual bars were in a different place each time. And if I redo that, you can now see they are individual elements.

So, we can see there still the same seasonality for Tyne and Wear, just not as much of the summer as there is in Greater Manchester. Greater Manchester goes from 200 to 235 to 255. Whereas Tyne and Wear, it goes up from 76 to 97,000, but then sort stabilises 102,000. So, when you have a limited number of items, then this can be good. It requires an even more limited number of items if it was a bar chart the other way, but I suppose it depends on how much space you’ve got. If you had it in individual page and has one of many visualisations, then it could be that you have more height than width. In which case, having it the other way around could work quite nicely.

Now, sometimes you really don’t care about how big the individual items are, it’s what is the percentage of the totality. So, here we can see everything is rising, maybe, but is going to Manchester rising get a greater rate than West Yorkshire or the West Midlands? Well, we can have a look at that using 100% stacked bar charts. What this does is this stretches the totality, so that it reaches the very top. So, right here in 2003, we don’t have that much to stretch. But going back to 1995, we need to nearly double the height to get it all the way to top. So, if I change this to 100% Stacked column chart, now you can see all of them being at compared in the same ratio. So, here we are looking less at the actual sales volume, but more at the percentage. So, looking at this we can see that it’s roughly the same throughout but there are variances. For instance, West Yorkshire goes down to 20%. Whereas previously it’s being in the 21%. It’s even getting to 22%. So, in 2011, it is the most affected, perhaps, of the entire six regions that we’ve got by the falling house prices. And falling house sales, particularly. So, this is probably the least used of the Bar charts. It’s there when you want to do comparisons between similar items, but you’re caring less about the absolute values, which you can see in the stacked column chart. And you just want to see ratio. So, here I can’t see the individual sales. I can’t see that sales went up between 1995 and 2003. Whereas here, I can see this quite easily.

So, we have got, now three different types of Bar charts and Column charts. We have got the Stacked variety, where they are on top of each other. We’ve got the Clustered variety. So, if I go down to a lower level, you can see how that can be used more effectively maybe. And we have got the 100% Stacked column charts, as well. So, here we can see if there’s any major variances in the seasonality. Looking at the formatting, you’ll see that there are no new categories of formatting to learn in all of these three. So, once you’ve mastered one, you’ll be able to find where all of the formatting is in all of.