I love it when the dataset for Makeover Monday is released early on Sunday; it feels a bit like Christmas coming early 😊. I haven’t had time to actively participate in Viz Review the last couple of weeks, so looking forward to joining the webinar this week and learning from the feedback.
This week’s dataset is looking at the cost of a night out in selected cities around the world; and below you’ll find the visualization referred to in the original article. I highly recommend actually reading the article because it offers additional insight on how the visualization is ‘interpreted’ and there’s lessons to draw from that as well.
Let’s take closer look.
What works well?
- Clear labels and colour legend.
- You can easily understand the ranking because the bar charts are sorted from most to least expensive.
- The Subtitle works well as it gives additional info and explains what the totals represent.
What could be done better?
- There’s too much clutter, the flags and image don’t serve any real purpose and are fighting for attention with the colourful bar charts.
- There’s no need for the axis labels as the bars already show the total per city and I found they added little value when trying to figure out the individual cost for each item just from looking at the chart.
- I can clearly see which city ranks highest based on the total, but I can’t really make out how the cities compare across the different categories.
- Comparing cities based on the cost of a few items doesn’t seem correct as it doesn’t take into consideration the actual cost of living. While Mexico City may seem like the cheapest option, it could still be an expensive night out for someone who lives there if you take into consideration their earnings and cost of living.
- Also, it doesn’t specify if this is the total cost for more then one person or not. You could make an assumption based on the cost of the Big Mac but it should be specifically mentioned on the chart so there are no misunderstandings.
- Update: I’ve been following the discussion on twitter and data.world and next to the original article provided with the dataset, folks also referred to the article posted by UBS which appears to contain additional information. Having said that, I do believe such information should be clearly mentioned on any visualization or table to provide proper context.
- And finally, I would also question the choice of a ‘Big Mac’ as the late-night snack of choice but that’s a different topic.
I went with a slightly different approach because not only did I want to see how cities ranked based on the total but also wanted to add some additional insight on how cities compared on the cost of the individual items.
Click on the image below for the link to the Viz on Tableau Public.
Thanks for reading!