0:04

I am now going to introduce a number of fundamental graphs.

Why do I do that?

Because I want to teach you first of all,

once you have identified attribute types,

what graphs are available to

visualize the combination of attributes that you have in your hands.

This is a preview of the graphs I'm going to talk about.

The bar chart, the scatter plot,

the matrix, the line chart and the symbol map.

Why do I do that? There are a number of

reasons but most importantly because they are widely adopted,

they are very effective and they are very useful.

A second reason is that they solve a very large percentage of visualization problems.

It's really surprising how many problems out there can be solved by

one of these five graphs or variants over these graphs.

And maybe even more important these are an amazing training ground

before you are able to discuss and design more sophisticated graphs.

So, I find that they are from the pedagogical standpoint extremely useful.

And as you will see they will allow us to reason

about graphical configurations and graphical communication.

So the graphs that I'm going to show you can all be

described as combinations of two or sometimes three attributes.

Let's start with the first one probably the most popular, the bar chart.

So before I start describing these charts let me tell you the charts that I'm going to

show are almost all based on

the vehicle collision data that we already used in previous lessons.

When I'm using a different dataset, I'm going to tell you.

So this first one bar chart,

and the bar chart allows you to visualize how

a quantity distributes across a set of categories.

So every bar represents one category and the length of the bar represents a quantity.

So it's a relationship between a set of

categories and measured quantities that refer to these categories.

In this specific example,

what I put here is the number of

vehicle collisions for different boroughs in New York City.

So you have boroughs on the x axis and the count of vehicle collisions on the y axis.

The next one is the line chart.

So the line chart allows you to visualize

how a quantity changes in relation to another quantity.

Where this quantity typically is time, not always.

The line chart can actually accommodate other information other than time,

but in general it's mostly used for time.

Okay. So in this specific case on the x axis we have

time and on the y axis we have the amount or count of vehicle collisions over time.

And as you can see we can identify some trends.

The next one is the scatter plot.

So, with the scatter plot we can visualize how a quantity relates to another quantity.

In this specific scatter plot I am visualizing information coming

from a food products dataset,

where every single dot represents one food product.

And on the x axis we have amount of calories and on

the y axis we have amount of carbohydrates.

And every single dot is placed in the position that represents

the value of calories and the values of carbohydrates that these data point has.

So scatter plot visualizes relationship between a quantity and quantity.

The next one is a matrix.

So in the matrix we are trying to visualize the relationship between two categories.

So we have one categorical attribute on

the y axis and another categorical attributes on the other axis.

Here in this specific example I

put in again data coming from the vehicle collisions dataset,

and in the vehicle collisions dataset we have information about

what vehicle types are involved in the collision.

So here what they put in this table is vehicle one and vehicle two in rows and columns.

And what you see within each cell is again the number of

collisions that happen in the combination of these two vehicle types.

The last one is the symbol map. It's a map.

And the goal of the symbol map is to visualize how

a quantity distributes across two spatial coordinates.

So, in a map we have latitude and longitude and every single dot or symbol is placed

in the location that

represents the data point or the object that is represented on this map.

In this specific case,

again we have the vehicle collision data and every single dot represents one zip code.

So every dot is a zip code and the size of

this symbol represents how many collisions

happened in the area represented by the zip code.

So once again, a symbol map visualizes how

a quantity distributes across two spatial coordinates.

So, this is a summary of the graphs

that I've just shown you and what kind of information,

what kind of attributes,

or combination of attributes can be represented with these graphs.

So, in a bar chart you have a combination between a category typically on the x,

but it can also be on the y axis, and a quantity.

So in this slide I also put the symbol O to identify ordinal data.

So when you see categorical C and ordinal O you know

that this graph can accommodate both categorical attributes and ordinal attributes.

Okay? So in a bar chart you can have on the x or y axis,

a categorical or ordinal attribute,

and a quantitative attribute,

that is the connection between the category

and the amount of that is calculated for this category.

In a scatter plot, you have a relationship

between two quantities, two quantitative attributes.

Every single dot represents one data item and

this data item is placed in the position that corresponds to the two quantities.

In a matrix visualization we have the combination of two categories,

of two categorical attributes.

Or if you want,

also between two ordinal attributes or even a combination of categorical and ordinal,

and the value that is represented within each cell is a quantity.

In a line chart we have typically on the x axis time or sometimes some other quantity,

so another quantitative attribute.

And on the y axis we have some quantities, some measurement.

So a quantitative attribute.

And finally, in the symbol map,

we have spatial coordinates and

symbols that are placed in the spatial coordinate of interest,

and the size of the symbol represents a quantity.

So we have space plus some quantity.