When used correctly, data visualisation is an incredible tool which can help us understand complex data and statistics. With the visual aid, we are more likely able to process and remember all of the information. When data visualisation is done wrong, it can make the wrong statistics stand out – or even change the message of the data completely. Here are 5 common mistakes that you might be making with data visualisation, and how to avoid them going forward.
Putting pie charts together incorrectly
A pie chart seems like a pretty simple piece of data visualisation. The size of each slice of the pie is decided by the percentage of the data – there’s nothing more to it, right? Well, wrong. The way that you set out your pie chart actually has a big impact on whether it makes sense to the viewer or not.
You should start with the largest chunk of the pie at the top, anchored at 12 o’clock on the circle and extending clockwise. The second largest slice should start from the same position and extend anti-clockwise. You can then add in the smaller segments.
The other option for a format is just to go clockwise from the top, smallest to largest.
Where possible, try to limit your pie to 5 segments. If there are more segments to represent, try a different type of graph. It’s harder to process the smaller segments when there are more of them, and it’s also tough to see just how large the majority is. When a slice of the data is less than 1% there’s just no way to show it fairly in the pie chart.
Dashed and dotted lines
Don’t include non-solid lines in your charts. They are distracting and can be misleading. Particularly where the gaps in the line fall, it can be tough to figure out the figures – and even more so if those gaps fall at a kink in the line. Instead, distinguish the lines using different colours which are far enough apart to be seen easily.
Scaling is a hard thing to work out correctly, but if you do it wrong, you could end up giving misleading data. For example, imagine that you are using a circle as a representation. When you double the size of the data for the next circle, you will also want to double the size of the circle. A common mistake here is to scale the circle by diameter – making it dwarf the other circle. However, when you scale by area, you will get a much more accurate representation.
Using colours on a heat map
On a heat map, putting bright colours in together will give a confusing message. Some will stand out more than others in a way that may not be representative of the data. Instead, choose one colour, and use different hues for the different data points.
3D bar charts look exciting. That’s a given. But what they also do is to skew perspective. Stick to simple 2D shapes, as these are easiest for the brain to digest. This will help the data to speak for itself, instead of giving out the wrong message. Don’t sacrifice readability for a flashy design. Keep this in mind with other design quirks, too.
Presenting data in graphs and charts should be done in the simplest and most intuitive way possible. When you stray away from that, you are likely to end up with the mistakes listed here – and more besides.
Amelia Knott is a team member at AuBiz.net– a free online ABN lookup tool. She is passionate about new marketing trends and branding strategies. She shares her insights through blogging.