Understanding how data visualisation can enhance communication and reports

Topic Progress:

Click on the numbered tabs to navigate this lesson.

(Image credit – Alberto Cairo)

As discussed throughout this course, we use data visualisations for a variety of reasons. To summarise, we visualise data because:

  1. It gives a clearer understanding of your data or story
  2. It amplifies the message that you want to get across
  3. It can aid quick decisions and analysis
  4. You can identify trends quickly
  5. You can share them easily with others

People remember images

If you have a key insight that you want to get across, visual communication not only helps to draw attention to it – it helps the audience remember it later. Think back to when you were learning maths at school. Did you find it easier to remember the equations that you were taught or the graphs that you drew?

The image above is the classic “normal distribution”, used in probability and statistics to indicate the spread of values in a certain type of observation. You should be familiar with this by now – but how well do you recall that the formula for a normal distribution is 

Y = { 1/[ σ * sqrt(2π) ] } *             e-(x – μ)2/2σ2

Basic design concepts

Designing good reports and presentations is balancing between clarity and communication. On the one hand, the clarity of a data visualisation is improved when we aim to reduce the number of design elements and focus on the information contained within. Colloquially, this is called improving the “data to ink ratio” – less ink, more data.

  • Simplicity: Choose a maximum of three colors and fonts and stick with them consistently.
  • Brevity: Keep the text short and to the point.
  • Creativity: Incorporate playful design that relates to the topic.
  • Two dimensions: Avoid 3D graphics: they distort data and look. 
  • Clarity: Label clearly, specify units, use a legend when necessary.


It’s not just about the chart

(What makes a visualisation memorable? – Borkin et al)

Conversely, however, a team of Harvard researchers tested over 2,000 visualisations on an audience and found that people are more likely to remember infographics if they have natural-appearing elements, such as illustrations or pictograms. More “clutter” led to better recall of a chart’s detail.

Every element that goes into your data visualisation is important. Annotations are carefully chosen to draw attention to important information, for example, and titles and icons are easy to recognise. The same team also found that readers generally spend a long time looking at titles and use these to interpret the chart information.

Ultimately, however, the complexity of a visualisation depends on your audience. Complex images may work well over a double-page spread in a magazine, but on a phone-screen sized website, aiming for clarity is best. 

As the famous designer Nigel Holmes puts it, as long as you understand “that the primary function is to convey statistics and respects that duty, then you can have fun (or be serious) with the image; that is, the form in which these statistics appear.”

Don’t forget colour

Remember that your choice of colours is as important as your choice of charts. If you are presenting data about deaths from a particular disease, or the spread of coronavirus infections, for example, think about how you want your audience to feel about the data. 

It might seem natural, for example, to choose red to illustrate the spread of disease, but that colour can alarm an audience because of the connotations it carries. A more thoughtful approach might be to use a muted palette in order to focus on the data, rather than the worry.

As the illustration above shows, there are some attributes to colour that are almost universally recognised. When including them in a chart, we have to understand how they are being interpreted.

It’s important to understand that your audience will have an innate understanding of some design elements too, and you should be mindful of whether to use this or avoid it depending on what you are trying to convey.