0:03

Now, I want to introduce you to a little exercise.

And the exercise is about trying to think

about alternate representations for these fundamental graphs,

that I've just shown you.

This exercise is useful because it starts teaching you how to reason in

terms of alternative graphical encoding strategies

for the same data set or for the same information.

This is very important skill.

Being able to think about alternative visual representations for the same information,

is one of the most important skills that you have to acquire.

And why is it important?

Well, it's important because it's important to

learn how to figure out what the alternatives are.

And also, because it's useful once you have a number of alternatives in

mind to be able to reason about which one is the most appropriate for your goal.

And in the next lesson,

we will cover mostly about how to reason on

the quality or effectiveness of a given visual encoding.

So now, I'm going to start with an example of

alternative designs for the information that

we typically have as an input to a bar chart.

So, in a bar chart, we typically have,

as we said before,

a number of categories,

and for each of these categories,

a quantity that is associated to this category.

So what I'm going to do next,

is I'm going to sketch a few alternatives,

and it's important for you to know that the goal of this exercise is not

necessarily to design or sketch visual representations that are good,

or even appropriate, is just to think about

what are other ways of representing the same information.

So, at this stage,

we are not concerned at all with whether a visual representation is appropriate.

The main goal here is just to exercise.

I would say the mental muscle,

that helps you in creating alternative designs for the same piece of information.

In a bar chart, as I said before,

we have typically this kind of information,

a number of categories like C_1, C_2, C_3.

And for each category,

we have a quantity like V_1, V_2, and V_3.

The bar chart has the three categories on the X,

but sometimes on the Y-axis.

And the length of the bar is proportional to the values that we have here.

So now the question is,

how else can we visually represent exactly the same information?

So, let me walk you through some possible alternatives.

So the first one is this.

3:47

So we have the three categories: C_1, C_2, and C_3.

And now we have to decide how to represent the three values.

Well, how about instead of using a bar,

we can use a simple dot.

So a dot will be here, here, and here.

So, that's a legitimate alternative representation.

But what is the problem with this representation?

Well, that it's hard for us typically when we see a cloud of dots like this one.

We don't necessarily interpret it as the values that correspond to these categories.

But if we draw a real line like this one,

all of a sudden,

we go back to interpreting this chart very similarly to the way we interpret a bar chart.

In fact, this is actually a very effective alternative,

and it's commonly called dot plot.

Very similar to a bar chart.

How else can we represent the same data or another representation?

Is this. We have category one,

category two, category three,

and we have dots like before,

but we connect these dots with lines,

so that will be the equivalent of a line chart.

With the main difference that on the X-axis,

we have categories, we don't have quantities.

Okay. What is the problem with that?

Well, the problem is this,

that when we have categorical attributes mapped to an axis,

and we visualize this information with the line chart.

The line chart tends to communicate to the reader the idea that these values are ordered.

Not only that, the other problem is that a line chart creates a very distinctive pattern,

and we tend to interpret this pattern as meaningful.

But when we have categorical information and categorical attribute on the X-axis,

this pattern is not really meaningful,

and these data items are not really ordered.

So, they look ordered,

but they are not ordered.

So, that's a very important problem that we have here.

That wouldn't be a particularly appropriate visual representation.

But again, as I said at the beginning,

the goal here is not necessarily to create effective visual representations,

but just interesting alternatives.

So, let's see, how can we visualize,

once again, the same information.

So let me create another axis.

I'll put the three categories here, once again.

So, how can we visually represent

the three values that correspond to the three categories?

Well, another way is to use area.

So, we have a few bubbles here.

So, the area of the bubbles represents the value that corresponds to each category.

So, what is the problem with this one?

Well, I think I can identify a couple of problems here.

So first of all, it's much less common to represent the data in this way,

so you can expect your reader to be less comfortable with this visual representation.

And the second problem,

as we will see later on in the course,

is that area size is not as effective in terms of

communicating quantitative information as comparing the length of two bars.

So comparing the size,

the area of two bubbles is not as effective

as comparing the length and the height of two bars.

Now, I want you to try to do exactly the same except view what the graphs are.

So this one is the scatter plot.

8:43

The next visual representation we are considering is the matrix.

So, the matrix accommodates two categorical attributes,

which are mapped into a few rows, and a few columns.

In this case, we have two rows and three columns.

So, let's call the two categorical attributes A and B.

C, here, to remind ourselves that these are categorical attributes.

And as you have seen before,

the values that are associated to the intersection of

these categories are represented like the area of a symbol.

In this case, the area of a square.

So, how does the information look like?

Though we have attribute A, which is categorical,

the attribute B that is categorical and an associated quantity.

So we have A_1,

B_1 in a certain quantity.

A_1, B_2 in another quantity.

A_1, B_3, then we

have A_2, B_1, A_2,B_2, A_2,B_3.

And for each of

this one we've a quantity.

Q_5, Q_6.

So, that's the kind of information that you have to use to

imagine different alternative visual representations for the matrix.

Now, select the main- First one,

is the symbol map.

So, in the symbol map,

we have a spatial configuration, something like this.

And we have two quantities that represent the location.

So let's use X and Y here.

And for each location,

we have, again, a quantity.

So, X_1,Y_1, X_2, Y_2, and so on.

And each one is represented with a small symbol,

and the size is proportional to the quantity.

So, now, your goal here is

to create alternative designs

for all of these graphs that I've just shown in the symbol map.

So, now it's up to you to design alternative designs for each of these graphs.