Week 49: The Cost of a Curry at Wetherspoon

Only a couple more weeks and we’ll be done with Makeover Monday 2017 edition. This year sure has flown by! Eva finally caught me up on the plans for 2018, which we’ll share on December 20. Sign up for the webinar here.

If you’re wondering why I chose “The Cost of a Curry at Wetherspoon” for the data set this week, it’s because I used it for a recent round of Data School interviews. The DS interviews are proving to be great test environments for Makeover Monday data sets. Anyway, onto this week’s lessons…

LESSON 1: SCALING BAR CHARTS

When communicating with bar charts, it’s absolutely essential that you scale includes zero. As Stephen Few wrote in “Effectively Communicating Numbers“:

This is because the lengths of the bars encode their values, but won’t do so accurately if those values don’t begin at zero.

Let’s look at an example similar to some of those I saw this week.

Notice here that the y-axis is truncated. If you visually compare the number of restaurants in the £26-£26.49 range to those in the £26.50-£26.99 range by looking at the length of the bars, you might think £26.50-£26.99 is about 20% the height of £26-£26.49. In fact, £26.50-£26.99 is about 50% of £26-£26.49 (12 vs. 25). If we properly start the y-axis at zero, we get a very different story.

If you really need to truncate the axis, consider a dot plot instead. Because dot plots are measured by their position rather than their length, the read won’t be mislead as their eye will need to go over to the y-axis to confirm the value.

LESSON 2: AVOID AUTOMATICALLY SIZING DASHBOARDS

Every week when we host Makeover Monday Viz Review, I mention that people need to stop using Automatic sizing on their dashboards. Why? Well, when you create it on your screen it might look great. But there’s no way for you to know the screen size of your audience. When I look at these vizzes on my screen, they often look very squished, making them difficult to understand.

My tip for you:

1. Start by setting the size to Automatic. This will fill up your screen.
2. Change the size to Fixed size. This will lock the screen into that perfect size in YOUR screen and will prevent resizing on anyone else’s screen.

Here’s a gif to show you how to do it.

LESSON 3: SUPPLEMENT THE DATA FOR ADDITIONAL CONTEXT

Yes, you’ve heard me preach before that I’d like you to stick to the data provided. What I really mean is that I want the data provided to be the main focus of your analysis. There are times when including additional data really helps accentuate the analysis and makes it more sticky. This week, Gina Reynolds did just that by including drive time from Big Ben so that she could see if that impact the price of various food items.

Nice work Gina! Such a beautiful viz!

LESSON 4: ENGAGE YOUR AUDIENCE

Getting your audience to interact and understand your visualization can be quite difficult. Anyone who’s been around a while knows just what I mean. One way to capture your audience is by making the viz personal to them. This week, Joanna Hemingway (one of my favorites at London TUG), created this interactive visualization to let us find the cheapest Wetherspoon nearby through a series of questions.

For me, Joanna’s viz demonstrates the method of engaging your audience extremely well.

LESSON 5: EFFECTIVE USE OF COLOR

Few weeks go by when we don’t talk about color in a recap. These range from using too many colors, to using the right colors, to using color effectively. This week, Kate Brown demonstrated how to use colors for highlighting.

Along with having a good title and excellent subtitle that describes the viz, Kate has purposefully used the color red for highlighting. This is a fantastically executed technique and one we can all learn from.

Once again the Makeover Monday Community continues to inspire Eva and me. You make writing these recaps and creating content for you fun for us. We appreciate it!

With that, here’s a gallery of my favorites from week 49.

FAVORITES

Author: Tori Colebourne
Link: Tableau Public

Author: Steve Fenn
Link: Tableau Public