Mastering Data Visualization: Theory and Foundations

The quickest interruption you'll ever see. Much less annoying than an unwanted phone call. I promise.

In this lesson I'll introduce the keystone work on human graphical perception. We'll learn that the ability of humans to decode graphs can be quantified.

In this lesson we will review the tasks of decoding position in a common scale, position in non-aligned scales, length, direction and angle. We will provide some real examples to get an idea of how difficult it is to perform each of these tasks.

In this lesson we will continue with some other (maybe more difficult?) tasks.

In this lesson we'll finally know which tasks are easy and accurate to do, and which tasks are the most difficult ones, meaning we must stay away from them.

Now that we know which tasks are best to use, we can question our current graph choices and modify them accordingly.

In this lesson you'll learn the must-follow rules of Data Visualization.

Sometimes a data visualization tries is built so that it exaggerates the effects present in the data. If this happens, we say there's graphical distortion. In this lesson we'll see some examples of that.

In this lesson we'll start the important chapter on Graphical Integrity. In other words, we will learn how graphs lie. In this particular lesson we'll learn that we can quantify how much a graph is lying and we'll learn to calculate that.

In this lesson it's your turn to calculate the lie factor from a plot. Let's see if you get it right!

Labeling and annotation can greatly improve your graph. In this lesson we'll discuss the main uses of annotation.

This is a crucial lecture. Do not miss this, because this mistake happens very often in graphs out there.

Another principle of Graphical Integrity talks about dimensions. In this lesson we'll learn how many dimensions our chart should have and how to count them.

Fancy a 3D graph? Well, I'm not sure you'll think the same after this lesson.

Let's talk about graphical sophistication. These tricks will help us achieve a better-looking, more efficient plot. In this lesson you'll learn to optimize the data-ink ratio to produce cleaner plots.

Do you know our eyes can perceive a lot of detailed graphical information? In this lesson you'll learn the concept of data density, and how to apply it to your plots.

Lastly, let's discuss proportion and scale. Should your plot be in horizontal mode? Or vertical maybe? What aspect ratio should you use? 

Have you ever heard the saying "correlation doesn't imply causality"? In this lesson we will discuss the difference between association, correlation and causality, and we'll explore the different explanations of why two variables might display a correlation.

In this lecture we'll discuss one problem that affects the data underlying many plots: selection bias and its "cousin" attrition bias.

When we plot some data, it's important to take into account all of the data that's important for our analysis. In this lesson we'll learn that quoting data out of its context can be tragic.

Sometimes we cannot extract conclusions from a plot because the data is not properly normalized. You have seen examples of this for sure, buy maybe you didn't realize it. Let's review this problem in this lesson.

Now let's review a famous effect: some data looks very different if we plot the aggregated data versus if we plot the different subgroups. In this lesson you'll learn why this happens.

In this lesson we'll review the classification of the different types of plots according to their purpose.

If you have multiple observations of the same variable and want to see how they look like, it looks like you need to plot a distribution. In this lesson we will discuss the many alternatives, how they work and how to use them.

If you want to compare several variables to see how they depend on one another, you'll need to use the plots we'll learn in this lesson.

The typical bar plot, the modern lollipop plot, and some other non-recommendable alternatives for displaying rankings are discussed in this lesson.

To compare part to the whole most of the people go for a Pie Chart. But honestly, in 2022, don't we have better alternatives to the Pie Chart? Luckily, we do. In this lesson we'll learn how to depict comparisons of part to whole properly.

Maps are a very specialized topic. In this lesson we'll review what we can draw by using maps and we'll learn some tricks from the best data visualization agencies.

This is a very, very common mistake. Still, people who draws graphs every day don't know when is it okay to cut the Y-axis and when is it mandatory to start at zero. After this lesson you know what to do in each situation.

The internet is full of line plots with a shaded area underneath. Actually, they look fantastic. But there is a big conceptual problem with shading an area plot.

Spaghetti charts happen very often when we draw line plots that are too cluttered. In this lesson you'll learn a few tricks to avoid them.

Have you ever seen a Dynamite Plot? I bet you have, but you didn't realize it. In this lesson we will review the purpose of error bars and we'll learn some alternatives to the dreadful Dynamite Plot.

Color is very, very important in Data Visualization. But, do you know what is the purpose of color? Do you know how to choose the right color palette?

Let's review the most common color mistakes so that we avoid committing a color crime!

CONGRATULATIONS!!! I'm so happy you made it! You should be so proud of yourself! Now, if you want to keep improving your Data Visualization skills, what should you do next?