Original subtitles

In this lesson.

We're going to continue to explore the different options we've got with the column graph.

So in the previous lesson, we saw how we were able to use our legend and to use our small multiples.

So what I'm actually going to do is I'm going to go back to just taking out our legend and we're going

to go back to using the region.

Just having three options here just makes it nice and easy to work with.

Now, when you displaying your data on a column graph, you may want to display it in a different format.

So instead of each item having a separate column, you may want to have one column and then just show

how much each item contributes to the total.

Now, this is actually called a stacked column graph.

Now, if we go across, you'll see that the little graph over here shows me that that is a stacked column

graph that's showing the item on top of each other.

So please note to make this work, you do need an item in your legend so that it's got something to

stack.

So when we select this now, you'll see now that it changes it to show me a stack version.

I do find that this graph is really useful for data analysis because it's showing me how much each item

is contributing towards the total.

And you can visually see this and you can actually see across the different manufacturers what the patterns

are.

Now, in this case, because we're using training data, the patterns tend to be quite uniform, but

you'll tend to find with your own data that it may change quite drastically and you can actually easily

pick up from these patterns where items are maybe not contributing as much to something as you would

expect.

So just something to look out for, something to try.

And you can see that we've still got our data labels on and the data labels are working pretty well

here.

I must say that they actually fitting in quite nicely.

Do give me a bit of useful information.

So as you can see now with our formatting, a lot of the formatting will stay exactly the way that it

was.

So your grid lines, your colors.

Actually, just one thing to note is that in your columns you do see that actually we've got different

colors here, so at any point you could come in here and you could change these colors that have been

used if you're not happy with them.

But as I said later on, we're going to be looking at themes.

And themes also allow us to change default color systems that we're using within power.

BI Right.

So that is actually our stacked column that we've that we're looking at there.

And as I said, we currently got the labels on.

If we turn them off, you'll see that you get that information and keep it on there.

It's working pretty well there.

The next part that we're going to be looking at is if we go back to our graphs again, you will see

that we get another type of graph which is actually called our 100% stacked column graph.

So what this is going to do now is it's going to keep the stacked column view.

But instead of having the values on the y axis, it's going to have percentages.

So we're going to see how much each item contributes to the total in terms of a percentage.

And this can be really useful again for your data analysis because it gives you a different view.

Now, at the moment, if we look at Northwind traders, to be honest, it's not really given as any

useful data because it's too small.

It's actually against the fabric cam.

It's not showing how much information that it can do.

But if we change this to a percentage, you'll see now that it's now showing us the percentage contribution

of each of the regions to the total.

So again, this is really just a different view of being able to look at our graph and being able to

understand it.

And as I say, you'll see now that you're getting a percentage contribution of each of the regions.

Okay.

What I'm going to do as well is we're just going to have another look at the way that our graphs work

together.

So, so far we've seen here that we've got our clustered column graph, we've got our stacked column

graph, and we've also got 100%.

What we also get though, is the ability to be able to do bar charts.

Now we're bar charts become quite useful is when you're actually analyzing quite a bit of of information.

So the bars are going to show you a horizontal view.

Okay, let's show an example of that.

So let's click over here.

Let's make a bit of space and let's create a clustered whoops.

Let's just change that one.

So I'm going to go back to this one being 100% go back to this one.

Let's create there we go.

So we created a new bar chart now.

And as I say, if you're looking at quite a bit of data, then the bar chart can be quite useful because

it's going to put the labels on the horizontal.

Let's say, for example, we look at our countries, so we pop that now onto our y axis because remember

it's a bar.

So it's going to go across this way.

And on our x axis, let's say we want to see our profit.

Okay.

So you can see now quite clearly the United States has got the majority of the profit.

But also these labels now are quite easy to read.

Whereas if this was on a column graph, you may find that it's not so easy to read from left to right

with all these labels.

So that's just something to look out for.

So we've got some of the profit by country.

And what you will find is that the stacked bar graph and the 100% bar graph work exactly the same way

as the clustered column graph does.

So I'm not going to take you through that.

You would you would just do the same things that we've done previously.

But one of the behaviors I do want to show you in this lesson is the ability to actually be able to

filter our graph from another graph.

And we've already seen this behavior in action When we were working with our tables, you may remember

we could choose an item on a table.

And what it did was it filtered the other visualizations that were on the page.

So this was called cross filtering.

So graphs can actually do the same.

So we've got two graphs here.

But just imagine that if we had more graphs, then you would be able to actually filter several graphs

at the same time.

You can actually filter several tables, you can filter matrices, cards.

It doesn't really matter which visualization it is.

The filtering works the same.

But we do get a little bit of a different behavior when it comes to graphs.

We get the ability to be able to do highlighting and filtering.

So let's see this in action.

So let's say, for example, I pick China.

Now what you will see is that immediately is that Asia is the only part that is highlighted here.

So what it is showing me is that with for China, how much it's contributing or the contributions for

each of the manufacturers.

So it's basically highlighting the data within this data.

So that is what filtering will do, is it will then highlight the data within the actual bigger graph,

if that makes sense.

So you can see the rest of the graph has still been shown.

Only the highlighted data is now being shown.

As I start to choose these.

You'll start to see.

That it's only shown the highlighted values.

And the way that that gets controlled is through the way that you control your interactions.

So if we go across to our formatting, you'll see that we get something called edit interactions.

When I turn on edit interactions, what it does is it allows me to take visualization and then determine

how the other visualizations are going to be controlled.

So because this visualization is now selected, what it shows me over here is these icons.

That shows me how the other visualization is going to be controlled.

So if I want nothing to happen, I pick none.

Now you'll see that when I select these items, nothing happens on this visualization.

If I pick this one, this is my highlighting.

So you'll see that as soon as I change here, it changes my highlighting.

And then the item.

And the last one is the ability to filter.

So if I select filter, you will now see that it changes to filter all the items.

Now, currently this graph doesn't make much sense when I'm filtering.

So what I'm going to do is let's just change that.

Let's go back to being a stack column graph.

So we'll go back to that view.

Let's select this again and we've now got the filter on.

So if we said United States, you'll see now only the United States figures will be done.

Go back to China and in the China.

So you can see that this graph now is being filtered versus this view was that it is only highlighting

within the actual graph itself.

So hopefully those makes sense.

It does take a little bit of time to get used to that behavior.

So I would suggest that this is something that you just practice with, experiment with, because also

when you're designing reports for your users, it's something that you've got to kind of consider.

Would your user actually understand the highlight or would they understand the filter better?

Or quite honestly, sometimes, are you just going to turn off that cross filter behavior?

Because sometimes it may confuse people.

But I'm going to leave it there for this lesson, and I will see you in the next one.