Andy and I are huge fans of simplicity and this week’s dataset was nice and simple, giving everyone a chance to stick to the basics if they want to and enjoy the nice (Northern Hemisphere) spring weather instead of spending many hours building vizzes. We asked you to analyze and visualize data about congestion in 10 European cities. More specifically, the data was about how many hours drivers spend stuck in traffic jams during their commute.
The original visualization was a bar chart and despite having room for improvement, it was simple and fairly easy to understand.
I love the creativity that so many of you bring to this project and that comes through in your weekly submissions. A makeover can be a completely new approach for a dataset or it can be simply an improvement of the existing chart. There is absolutely nothing wrong with creating a bar chart that presents the data more effectively and gives your audience a quicker understanding of a topic.
Don’t feel like you need to go all out every week and come up with a groundbreaking viz. Bar charts are fine and often quite sufficient.
Welcome newbies!
It’s always exciting to have new people join Makeover Monday and contribute their ideas and vizzes to the project. We love having you all become part of our community and hope you’ll find the project helpful, the weekly exchanges, recaps and webinars useful for your development and that you enjoy getting to know this online community.
If you’re unsure how all of this Makeover Monday stuff works, here is something to get you started:
- Check out our Makeover Monday 101 webinar
- Find our weekly recap blogs here
- Check out the Gallery where we feature the weekly favorites and mentioned authors’ vizzes to showcase the creativity and diversity of this community and the work people produce
- Follow our project on Data.World where we publish the weekly datasets. It’s there that you add your submission in the general discussion and you can add it to the Viz Review discussion if you want live feedback during the weekly webinars
- Register for upcoming Viz Review webinars to get feedback on your submissions
- And don’t worry about getting stuff wrong. While we may not be able to hand-hold everyone through the process, we have a great community who is happy to help you, so just jump right in and get started… find the data here
LESSON 1: ASSUMPTIONS CAN BE DANGEROUS
In this week’s dataset there were 10 European cities. The background article had the title ‘Which European commuters spend the most time in traffic jams?’ which led to numerous people stating that these were the 10 worst cities for commuting or the top 10 most congested European cities.
They’re not.
They’re just a selection of cities which EuroNews chose to visualize. Why those 10? I don’t know, but what I do know is that when you go back to the original data source, there are cities in the actual top 10 which are not in this list.
Be careful with your assumptions and the statements you include in your visualizations. Do a little extra background research, at least read the article that comes with the data. This is bread and butter stuff for analysts. You don’t have to fact-check every sentence, but if the original article doesn’t mention that these are the top 10 most congested cities, then I’d suggest to not make that claim, especially when it can so easily be dismantled when looking at the original data.
LESSON 2: TREEMAPS SHOULD REPRESENT THE TOTAL DIVIDED INTO ITS PARTS
We saw quite a few treemaps during this week’s Viz Review. Treemaps are designed to show the components that makeup the whole, i.e. the entire treemap is 100% of something and the individual rectangles that comprise the treemap add up to 100%.
Treemaps can be used when your data contains all the components that makeup the whole. For example, you could have the total sales for a year with the components being ALL the different regions and countries your organization operates in or ALL the different product categories that structure your product hierarchy.
In these cases the treemap shows the individual regional and country or category contributions to total sales and we’re good!
Aparna Shastry submitted the following viz for Viz Review and our feedback was that a treemap isn’t a good way to present this data because the dataset doesn’t add up to a whole but rather contains 10 selected cities.
She went on to create a completely new visualization, using a bar chart to compare the cities and also added a number of annotations to share her insights.
It’s great to see when our feedback appears to be helpful for people and of course to come across a viz that presents the data and insights more effectively. Well done!
As I was travelling this week, I couldn’t engage much on Twitter but enjoyed seeing all the submissions on Data.World, so here are my favorites from this week…
FAVORITES
Author: Michael Mixon
Link: Tableau Public
Author: Tushar More
Link: Tableau Public
Author: Staticum
Link: Tableau Public
Author: Ali Motion
Link: Tableau Public
Author: Sarah Bartlett
Link: Tableau Public
Author: Heqing Huang
Link: Tableau Public