Dos and don’ts for presenting data ethically

By Apolitical

Thursday March 4, 2021

A man with a remote control slide presenting data
Strive to make your data presentation as accurate and honest as possible. (Image: Adobe/Ratchapon)

We know by now that data presentations can enhance people’s understanding of complex information, but it’s important to acknowledge the ways that they can mislead audiences, too.

Hiding relevant numbers, presenting too much information, or choosing the wrong kind of graph risks unintentionally distorting the message in your data.

And while making mistakes is an inevitable part of learning how to visualise data, there are some ground rules you can follow to help make your presentation as accurate and honest as possible.

Alberto Cairo, Knight Chair in Visual Journalism at the University of Miami and the author of How Charts Lie (2019), shared some tips for ensuring that your visualisations are ethically sound. We’ve listed four rules below:

Don’t let confirmation bias dictate your data choices

“Humans are confirmation machines,” said Cairo. “Without knowing it, we can easily become laser focused on finding only the numbers that support what we believe.”

When we want an idea to be true, we tend to seek evidence that supports our beliefs and interpret it in a way that confirms these beliefs. For Cairo, being aware of this fact is a good place to start when you’re thinking about conveying data to an audience.

Creating a story or a visualisation out of data “can easily devolve into a falsification process,” warned Cairo. “If you only seek the numbers that support your argument and nothing beyond that”.

So how can we counteract this? “You must train your brain to do the opposite: to expand your scope and seek out information that complicates your ideas,” Cairo said. “Numeracy is like riding a bike: practice paying more attention and it’ll become second nature eventually.”

Don’t conceal uncertainty

Uncertainty is an inherent fixture of life. Don’t deprive your audience of nuance, and be sure to include data points that contradict your argument, if those exist, said Cairo.

“Uncertainty is essential, and you cannot hide it. If you withhold information that casts uncertainty on your argument,” he said, “you are manipulating the existing data to create a narrow picture of what’s happening.”

This is also about creating a space for those stimulating conversations that may arise from uncertainty: “A good chart may help you to answer a question … but they can also be great for piquing our curiosity and prompting us to ask more questions.”

Revealing contradictory or confusing data points also shows integrity and can increase your credibility in the eyes of your audience.

So, don’t be discouraged if your message is complicated by ambiguous data — embrace it as a way both to spark conversation and establish trust with your audience.

However, as you learned in the previous section, it’s still important not to overload the audience with extraneous information. Use your best judgement and ask yourself: “if I was listening to this presentation, would excluding this information help or hinder my understanding of the core issue?”

Do contextualise your data visualisation

To mitigate potential misunderstandings, contextualise the data that you’re presenting thoroughly.

Not only will this minimise the chance of your audience misinterpreting your data, but it will also help you to keep your own confirmation bias in check.

“Lay out your case step by step,” said Cairo, “and attach each link of your reasoning chain to preceding and subsequent ones”. Do this and “it’ll be much harder to make unfounded and deceptive claims.”

One effective way of providing context is by annotating your charts clearly. “A chart can be a visual argument but is rarely sufficient on its own — I am a great believer in combining words and visuals”.

Cairo suggests annotating your charts with the assumption that every audience member will misinterpret them: this will force you to provide even clearer titles, disclaimers and context than you might think to.

“Clarify the origins of the data and offer possible alternative interpretations when you can,” he said. “This is an essential practice when you’re presenting data — we want information to be presented simply, but never at the expense of understanding.”

Do use reliable data

It might seem obvious, but it always bears repeating: use data from trusted and credible sources. “For a chart to be trustworthy,” Cairo said, “it must be based on reliable data.”

There are a few ways to ensure that your data is accurate. One of the easiest is to make sure that your data is coming from a primary source. If you find data you’d like to use, track down the original source and verify that it is legitimate.

Ask under what conditions the data you’re using was generated. Was it published by an organisation with a political agenda? Is it biased? Use discretion if the answer is yes.

Another way to increase your presentation’s reliability is to use the most recently published data available: if it’s older than a year or two, make this clear in your presentation.

And finally, “if you are unsure about your process, collaborate with experts! Especially if you are making your data-driven argument public,” Cairo advised. “Never, ever publish something that hasn’t been looked at by an expert.”

This article is curated from Apolitical.


We can see the male gender bias of all-boys schools by the books they study in English

Who is (probably) today’s best male tennis player?


About the author
Inline Feedbacks
View all comments
The Mandarin Premium

Insights & analysis that matter to you

Subscribe for only $5 a week


Get Premium Today