top of page

Beware of Deception When Using Shapes for Data Visualization


There is skill (and potential deception) when we use shapes for data visualization of one-dimensional information in two dimensions. Are you Communicating with FINESSE?
There is skill (and potential deception) when using shapes for data visualization of one-dimensional information in two dimensions.

Fortune gets it right! In its annual publication of Fortune 500 companies, Fortune correctly used two-dimensional shapes (circles) to depict one-dimensional values such as annual revenues. Using shapes to distort and deceive comparisons is an old trick frequently used for manipulation by one side of an argument. We’ll discuss this data visualization trick in this brief article.


The Trick

Suppose your total revenues were $100 per month, and your leading competitor had total revenues of $10 per month. Clearly, you are ten times stronger from a total revenue perspective from your competitor. A one-dimensional graph, such as a bar chart, makes an honest visual representation of the difference.


Now represent the one-dimensional value as a two-dimensional object. In this case, let’s use a square. I could use the one-dimensional values, $100 and $10, as the length of the square. If I erroneously enter the one-dimensional value for one side of the two-dimensional shape, then I calculate areas of 10,000 and 100. That distorts the difference between the larger and smaller company to 1000 times, not 100 times.

Transferring one-dimensional data into two dimensions can be tricky.
Transferring one-dimensional data into two dimensions can be tricky (and deceptive).

If you are trying to mislead, deceive, or manipulate your audience, simply use this erroneous visualization. Showing it fast in a presentation also helps because the audience does not have time to think much about what they just saw.


The Correct Way

The correct way to display a one-dimensional object in two dimensions is to calculate an area based on the one-dimensional value. So, with a one-dimensional value of 100, the side dimension of your square would be 10. For the second company with a one-dimensional value of 10, the side dimension would be 1.

Getting the two-dimensional depiction and using a multivariate visualization are tow good practices.
Getting the two-dimensional depiction and using a multivariate visualization are tow good practices.

If the second company would grow to half the revenue of the larger one, say $50 per month, the side dimension of the square would be 7.071. The resulting visualization accurately represents company A being twice as large as company B.


Fortune Gets the Data Visualization Correct

I quickly checked the great chart by Nicolas Rapp on page F28 in the June-July issue of Fortune. The two scenarios from the chart that I examined were 1. Exxon Mobile compared to Chevron and Marathon Oil, and 2. Walmart compared to Amazon and Costco.

Bar charts are always correct for one-dimensional data. Make sure the bar widths are always the same.
Bar charts are always correct for one-dimensional data. Make sure the bar widths are always the same.

Using the one-dimensional revenue data provided in the edition and taking screenshots of the images, I exported the information into one of my Excel-based tools for a quick check. Guess what? The visualizations had been developed correctly.


Common or Uncommon?

I review and see a lot of visuals in my work. I notice the misrepresentation of one-dimensional information in two dimensions a couple of times per year.

The bigger question is whether it is done intentionally or not. My anecdotal experience is that it is intentionally done about half the time. I saw "about" because no one I ask ever admits they created a deceptive graphic.

Fortune used the correct technique (right) for depicting one dimensional data in two dimensions.
Fortune used the correct technique (right) for depicting one dimensional data in two dimensions.

How often do I see this misrepresentation in the media or on social media? Frequently, but I cannot tell the difference between intentions and incompetence without being able to question the content creator.



Communicating with FINESSE

Shapes can be deceiving. As a trusted advisor, be careful when using a two-dimensional shape to depict a single-dimension measurement such as dollars, number of occurrences, height, or weight. Use a bar chart with a fixed-width bar as a better, less deceiving option.


The I in FINESSE stands for Illustrations. Seek to understand how poor data visualization impacts your information (and credibility).


The E in FINESSE stands for Ethics. Treat others the way you wish to be treated.


 

Communicating with FINESSE is a not-for-profit community of technical professionals dedicated to being highly effective communicators and facilitators. Learn more about our publications, webinars, and workshops. Join the community for free.

 

Comments


bottom of page