data visualisation examples
December 7, 2019

More great Data Visualisation examples from this years awards

By Paul Laughlin

Let’s look at 3 more great examples of Data Visualisation award winners from the latest Information is Beautiful awards.

As I mentioned in my first post reviewing this event, this is a great opportunity to be inspired in your own practice. For creative examples and elements of winning Data Visualisations to inspire improvements to your own charts. As I advice on my Data Viz training, you hone your craft by creative experimentation & testing.

In our first post on this event, I shared examples from the following categories:

  • Arts, Entertainment & Culture
  • News & Current Airs
  • Maps, Places & Spaces

Data Visualisation examples can inspire 3 dimensions of change

This time I will share 3 more examples that I believe could help inspire you. I have focussed on those with elements that are relevant for everyday Data Visualisers. By that I mean those analysts or data scientists for whom data visualisation is only part of their job.

Even if you are limited to using Excel, there should be ideas to inspire you here.

This week I spent a very enjoyable day with one of my clients, reviewing their progress with data visualisation. As they presented examples of their improvements 3 themes struck me. These 3 common themes covered a lot of the reasons for improved data viz effectiveness:

  1. Decluttering & simplification of charts
  2. Attention to detail to ensure each element is clear (inc. fonts)
  3. Incremental improvement (step by step & “sweat the small stuff”)

Considering all that, here are 3 other award winners I hope inspire your progress.

Going Grey = profiling Japan

The first example I want to share is the bronze winner in the People, Language & Identity category. One of a number of impressive entries from the Data Viz team at Reuters.

My reason for sharing this example is prompted by a concern that is shared on most of my Data Visualisation training days. That concern is how to help viewers who might have colour blindness. We have talked over a number of potential approaches, but one is to not rely on colour at all.

This engaging and well written data journalism piece makes extensive use of grayscale (and some hashing) to do just that. It’s true that the reason for use of grey is stylistic, in line with content profiling ageing Japanese population. However, it still provides some good examples.

I encourage you to review how greyscale is here used effectively in line, violin & area charts, as well as maps. Could you use this approach more, potentially with one accent calendar for preattentive ‘pop out’?

Going Gray

The world’s population is getting older. Japan is on the forefront of this demographic trend that will affect Germany, China and Italy in coming years.

The Millions who left East Germany

Next, I share this Silver winner from the Politics & Global category. It comes from Zeit Online, a German newspaper whose Data Viz team are regular winners at these awards.

I recognise that once again this is a data journalism article with scrolling & interactive data viz features that you may not be able to replicate with your current tools. However, if you scroll down through this post, you will see a number of basic charts that I recommend inspire you.

Effective use if made of a line chart with annotations close to the data points. This also uses the trick of shading specific periods of time to highlight the different temporal & contextual reasons for changes in the time series. Uses these overlays in order to step reader through the chronology can allow for real depth to be added.

This overview is then drilled down into some appropriate use of small multiples (area charts), interactive maps & scatter plots. Consistent meanings to colour hues for categories also enable the viewer to navigate between charts.

A useful example of when to use a range of basic charts that are available even to those only using Excel:

East-West Exodus: The Millions Who Left

Lesen Sie diesen Text auf Deutsch On the night of Oct. 3, 1990, German dignitaries sang the national anthem on the balcony of the Reichstag in Berlin. In the surrounding streets, people celebrated with German flags and bottles of sparkling wine.

Model Tuning and Bias-Variance Tradeoff

My final selection for this data visualisation post, is the bronze winner from the Science & Technology category. Part 2 of an entry of a ‘Visual introduction to Machine Learning‘ from Stephanie Yee and Tony Chufor of R2D3.

This example appealed for a couple of reasons. Firstly, it’s content is very relevant to our readers. The visualisers helpfully step viewers through each step of setting parameters for a model, adding complexity and then refining to avoid over fitting or bias. It is a great tutorial in the statistical checks every modeller should consider.

In addition this is an example of how to use a variety of creative charts to visualise complexity. For many years AI & predictive modelling were limited in business applications because of their perceived “black box” nature. A key challenge for Data Science leaders is to shed light on models, algorithms and why leaders should be confident in their ‘intelligence’,

In this post, Stephanie & Tony make very effective use of a series of unit, sunburst & scatter plots. Even including venn diagrams & pie charts! Well worth looking at how they take on the challenge of visualising the dynamics of model fitting and testing.

How could use even part of their approach to bring your key statistical models to life for a non-technical audience?

A visual introduction to machine learning, Part II

A visual introduction to machine learning-Part II The goal of modeling is to approximate real-life situations by identifying and encoding patterns in data. Models make mistakes if those patterns are overly simple or overly complex. In Part 1, we created a model that distinguishes homes in San Francisco from those in New York.

What next for your data visualisation examples?

I hope those examples intrigued you and got your creative juices going.

But, what I would most want to do is help you improve your own data visualisation practice. So, after reading this post, I encourage you to make some notes. Identify at least one thing you can put into practice in the next two weeks.

If you do achieve a significant chart makeover as a result of any of our posts, please send me a copy. I’d love to interview you about how & why you improved your chart and the benefits achieved as a result.