Browse by category


Blog archive

2019October 2019 (2)September 2019 (2)July 2019 (3)June 2019 (2)May 2019 (1)April 2019 (2)March 2019 (1)February 2019 (2)January 2019 (2)2018December 2018 (2)November 2018 (1)October 2018 (2)September 2018 (1)August 2018 (2)July 2018 (1)June 2018 (1)May 2018 (1)April 2018 (1)March 2018 (2)February 2018 (2)January 2018 (4)2017December 2017 (6)November 2017 (4)October 2017 (5)September 2017 (3)August 2017 (3)July 2017 (3)June 2017 (1)May 2017 (2)April 2017 (1)March 2017 (1)February 2017 (2)January 2017 (2)2016December 2016 (2)November 2016 (1)October 2016 (1)September 2016 (1)August 2016 (1)July 2016 (1)2015December 2015 (1)January 2015 (1)2013December 2013 (1)2012November 2012 (1)

 

While Mat has been engrossed in virtual reality, I have fallen down a less glamourous (but in my opinion, just as cool) rabbit hole of data visualisation. While that particular rabbit hole may or may not end up bearing fruit, to thoroughly mix my metaphors, it has caused me to reflect on the way that - with creativity and the right tools - data and statistics can engage, inspire and educate people or be a dangerous tool for misinformation.

 

Awesome data communicators 

The best data analysts are the ones who are also fantastic communicators. I'm a big fan of Mona Chalabi, the Data Editor at The Guardian US. She creates hand-illustrated data visualisations, often focussing on gender and racial inequality. Here are a few examples:

Depression and anxiety data Gendered, genderless and Changes in women's pay 1987 - 2017

Who said maths can't be creative? (Admittedly statistics are very often creative but in other ways).

The late Hans Rosling is another well known, popular data analyst. The Swedish academic developed software that shows trends in data over time. He recorded a number of TED talks and BBC documentaries where he brought to life figures from the UN and World Bank, debunking myths and preconceived ideas about health, poverty and demographics in the so-called "developing world".

 

Gapminder, the foundation that Hans started with his son Ola and daughter in law Anna, continues to produce videos that make the world understandable through statistics. Google bought Trendylzer, the technology behind Gapminder, and launched a public data explorer so you can create and share charts generated from official data sources.

 

Unreliable data

Of course data can mislead as well as inform. Tyler Vigen's Spurious Correlations demonstrates perfectly that correlation does not equal causation with his simple graphs of apparently similar trends:

 

Age of Miss America vs death by steam graph

 

It's important that any kind of graph, chart or data visualisation should use and cite reliable and relevant data sources. Inaccurate, incomplete, outdated or misused data can have serious real-world consequences when it forms the basis of policy or decision-making.

 

Picking the right chart

Assuming your numbers are correct, the next step is to find the best way to represent them.

People ❤️ pie charts. In certain circumstances they are fine, but in many others they are confusing or misleading as this post about some of the problems with pie charts explains. Apart from being a design nightmare in terms of colour combinations, the pie chart below is difficult to read because there are so many slices and it's impossible to tell the difference between similarly sized slices.

A plain old bar chart would be much better in this scenario.

Pie chart of US populations by state

Even relatively simple charts can become more interesting and informative with interactive or animated elements. Google has a gallery of charts available for data visualisation. 

 

¡Animo!

Animated charts are particularly effective when showing trends over time, for example in this BBC article about the effects of climate change in 7 graphs.

Here's an example of an animated graph showing Mat's Fitbit activity and sleep data over the last 6 months (of course he records all his data!) Click the button to see the charts animate between sleep and activity data. This chart was made using Google's animated chart library.

It takes up less space on the screen to show both data sets on the same graph and it's more interesting for viewers to toggle between the figures.

 

Conclusion

People don't generally get excited about lists of numbers but they do like pretty pictures and moving things. Used well, data visualisations can be an engaging way of telling a story or illustrating a point. They can fun and funny, hard-hitting or inspirational. As long as data is accurate, relevant, up-to-date and clearly presented, graphs and charts can be powerful tools - just think twice before using a pie chart.

Tagged under: Data