Data Visualization 101

Hi friend,

Good to know that you want to learn about the fundamentals of data visualization. But before getting that let's just define the two terms data and visualization.

data : facts and statistics collected together for reference or analysis.

"there is very little data available"

visual : relating to seeing or sight.

"visual perception"

The What, The Why, The How

What is data visualization ?

Using the above definitions we can understand that data visualization is the representation of facts and statistics in a form that can be processed by our sight.

Why do we need to visualize data?

The human population's excessive use of technology leads to the generation of a lot of data about the human population.(To be specific, the world creates about 2.5 quintillion bytes of data in day). It would really wise to use this data to make better decisions. To make better decisions with data, we need to understand the data. We visualize the the data to understand it and get the underlying insight it has to offer.

How do we visualize our data?

There are two parts to visualizing our data. First of all we need know what we are trying to measure (metrics). Secondly we need to figure out how we want to present that data (user interface). We will mainly focus on the presentation aspect in this article.

In terms of the technical requirements, there are many ways to visualize data. We can use a visualization tool like power BI or tableau, we can use programming languages like python or R. There are few other ways too.

Now let's switch gears and understand the different factors that affect the way our brains perceive an image.

Contrast and Patterns

Let's take a look at 2000 words. (A picture is worth a thousand words)

Here we only decipher a tree landscape. Right? (That's because there's no contrast)

Now we can see a bear inside the tree landscape. (Changing colors created contrast in the image)

This shows us how our brains process contrast. But there's more.

Our brains are continuously looking for patterns in the new information around us (pre-attentive processing). Because of this it is easier to detect the differences around us.(specially noticeable in patterns).

Here are a few illustrations of how our brains use pre-attentive processing to find differences.

The rectangle with a different color stands out.

The larger rectangle stands out.

The rectangle with a vertical orientation stands out.

Using the above principles in data visualization

Understand :

1. our eyes don't process images in a particular order. (Unlike the way we process text i.e. from left to right, and up to down)
2. Our eyes focuses on what stands out.
3. Our eyes can only handle a few things at once.
4. We look for meaning in data.
5. We are guided by cultural convention.

example :

1. The steep rise in the graph stands out.
2. The focal message is that the incarceration rate in the US increased after 1970.

bad example :

When there are more than five variables, then our eyes process the whole image as one single whole. Therefore there is no clear message in this chart.

bad example :

It appears as if the participants represented by orange dots are the top performers. But when we look closely we see that that's not the case. Therefore when using more than one color, then assign deliberately.

bad example :

This chart looks confusing because the days of the week are on the y axis. It would have helped to place it on the x axis. Social conventions are important my friend.

I hope this was helpful.

This article was heavily influenced by this youtube video (Do check it out):

Thank You