Can infographics tell the truth about the dangers of a dog’s breed?

We all know that we can lie using infographics. This can be achieved in a variety of ways including choosing to include only data that is favourable to your argument, omitting relevant data and presenting data in a way that can be interpreted differently to what it truly represents. The internet is full of such examples.

But if we can lie using infographics, the mathematical consequence of this statement is that we can also tell the truth using infographics!

Here is an example

Pit Bull attacks are reported by the press with some degree of regularity. In most cases the reports describe Pit Bulls, especially the American Pit Bull Terrier, as a highly aggressive breed of dog, if not the most aggressive dog breed. A quick search on the internet will reveal several news articles affirming that the Pit Bull is the most aggressive breed. Such statements have been repeated so much that it is widely believed to be correct, regardless of whether it is true or not.

Which dog’s breeds are really more aggressive?

Such statements are not indicative of reality. The American Pit Bull Terrier is not the first breed in the “aggression ranking”. It’s not the second – not even the third. Can you guess what the most aggressive dog breed is? It is the Dachshund. You may be understandably surprised and confused but the explanation is quite simple if we can correctly define what aggression actually is. The Oxford dictionary defines aggression as a “feeling of anger or antipathy resulting in hostile or violent behaviour; readiness to attack or confront”. If you consider all the Dachshunds or the Chihuahuas you’ve met, you may recall that they barked at you and showed very hostile behaviour. But we tend to see these attacks funny and even cute, because what harm, if any, can a Chihuahua do to you? Perhaps a small wound on the tip of your little finger? Certainly nothing more serious than that. Because of their small size, we simply don’t worry about a Dachshund or a Chihuahua barking and as a consequence we normally fail to understand that these small breeds are by far the most aggressive dog breeds.

Strength VS aggression

The obvious difference between a Dachshund and a Pit Bull is their size and the strength of their bites. A Dachshund’s bite offers little or no harm. On the other hand, a Pit Bull’s bite can be very serious or even lethal. Just like the bite from the most loved breed, the Labrador: when a Lab decides to bite (and sometimes they do), significant damage can be done. That being said, it turns out that a Pit Bull is, on average, significantly less aggressive than a Dachshund, but a lot more dangerous than a Dachshund if it decides to bite you.

Here is an infographic closer to the truth

So, the widely accepted concept that Pit Bulls or Rottweilers are the most aggressive dogs is simply wrong. I hope I managed to convey this to you in the previous paragraphs but we all know that a visual aid can help us to better understand the concept. Can a visualisation help us to tell the truth about Pit Bulls? The following is one attempt to do so:

Dogs copy


Note that the danger ranking is a result of combining aggressiveness with size. Maybe using size is not the best option (Dobermans are bigger and more aggressive than Pit Bulls, but are responsible to fewer deaths): If we had used strength of the bite instead of size, the chart possibly would be even more truthful. What’s more, it would be more truthful still had I corrected the danger ranking according to the frequency of each breed, which I didn’t (a given breed can be low on the statistics of fatal attacks simply because that breed is rare).

Misleading visualisations or deceptive infographics can attract more clicks and can be appealing. It is, however, important to understand any bias used. Use visualisation to tell the truth, or, if impossible, create a narrative that reflects the truth as closely as possible. People will realize you are providing accurate information and will associate you with being a reliable source of information.

The Pros and Cons of Using Maps to Visualise Data


By Gerardo Furtado.

We all love maps. They provide orientation, they give us a context for the data we are using, and they are beautiful to look at. Therefore, if the infographic we’re planning to design includes geographic data, we’ll usually put a map somewhere in the design. However, here’s a word of warning: In some situations, maps add nothing to the reader’s ability to understand the data being presented. In fact, in some cases, maps can be misleading and deceptive—and lead the reader to the wrong conclusions. In the former scenario, maps should not be used. In the latter, they cannot be used!

Let’s start with an example of a map that adds very little or no value to the reader. Suppose I’ve designed an infographic about the GDP of European countries. It would make perfect sense to use a map to illustrate my data, correct? Maybe not. Here’s a very simple map that I made using Tableau Public. I used a filled map with a sequential colour scheme; the GDP of each country is shown in thousands of euros:


But what value does this map add for the reader? Does it really help the reader understand the data? It’s hard to compare, classify, or rank the individual countries; there is nothing to distinguish one from the other. The reader’s eyes have to move from country to country, take note of a value, remember that value, and then take note of the value of another country; by that time, the reader doesn’t remember the previous value anymore! So, if you look at it from the reader’s point of view, you’ll conclude that the person who made this graphic (that is, me!) used a map just for the sake of using a map. You know where Germany is, you know where the UK is, and you know where Italy is! You don’t need a map for that!

Let’s display the very same data in a simpler form: a good, old-fashioned bar chart:


The bar chart allows the reader to make much more sense of the data than the map did. The reader can easily see each country’s ranking (Germany is first, and Malta is at the bottom). The reader can compare the countries and easily tell which ones are above average, which are under average, and so on. We don’t have a map anymore, and the map was an embellishment to the graphic, but was it worth it? Did it add any value for the reader? This is an example of a situation where a map added little or nothing to the reader being able to understand the data, and it did no harm to get rid of it.

What about situations where a map is actually misleading and causes the reader to draw the wrong conclusions about the data? Many examples of this can be found on the Internet, and most of them occur because of one simple fact: Many geographic heatmaps are really just population maps! Let’s look at an example: I’ve obtained the GSP (gross state product) of every U.S. state and, using Tableau Public again, I’ve displayed them on a heatmap, where dark red indicates a state with a very low GSP and dark green a state with a very high GSP. Here is the map, with GSP in millions of U.S. dollars:


Here are the conclusions drawn from looking at this heatmap: California, New York, and Texas have high GSPs, followed by Florida and Illinois. Then, there is a void in the Western and Midwest states, which have very low GSPs. But wait a minute: Why is the GSP higher in California, New York, and Texas? I’ll give you a hint: Do you know which three states have the highest populations? California, New York, and Texas! This “GSP map” is actually a population density map! The more people who live in a state, the higher that state’s GSP. How could we design a clearer map? The first, most basic step, which students learn in Statistical Analysis 101, is to correct the data by adjusting it according to population. In this case, we’ll use GSP per capita. Here’s the result (again, with GSP in U.S. dollars):


Wow! This map looks very different! There is no longer a big spread between the East and West coasts; they are more equal, more similar. We see Wyoming emerging in the middle of the map, and the East Coast states, from Delaware to Massachusetts, have a high GSP per capita. Now, we have an actual GSP map … not a population density map. This map adds value to the reader and helps them visualize and understand the data.

Perhaps the best way to sum it all up is with this joke from