Visualizing 5-year Cancer Survival Rates Through Sparklines

Last weekend I discovered the Makeover Monday challenge when I saw Andy Kriebel and Eva Murray’s new book, #MakeoverMonday, trending on my Goodreads feed. Inspired by the book and challenge I am committing to do at least one #MakeoverMonday viz per month. See more about my first below!

Makeover Monday 2018 Week 41 – 5-year Cancer Survival Rates in the USA:

Original Viz on Left from Our World In Data, #MakeoverMonday Viz on Right (click image to view the interactive dashboard)

Creating My #MakeoverMonday Solution:

Step 1 – Analyze:  When analyzing the original chart from Our World in Data, there were a few features that worked and didn’t work.

What worked:

  • Chart selection: the dumbbell chart selection did an excellent job to show the change between two points visually
  • Visual cue for direction: the arrows within the dumbbell chart acted as a great visual cue to help the reader quickly see the direction of change
  • Title location: the choice to put cancer type as a label on the chart vs. a y-axis made the chart easier to read and more visually appealing

What didn’t work:

  • The color choices: there was not much contrast between the red, blue, and pink colors causing them to add more noise than value
  • The sorting: the graph was sorted from top to bottom by the maximum cancer survival rate (either 1977 and 2013) instead of the cancer survival rate in 2013. I felt this hid an interesting finding which was that the survival rate for Cervix Uteri and Uterus cancer declined since 1977
  • One showing two points in time: after looking at the data behind the original chart I felt the authors missed the entire story by only showing the change in survival rate between two points in time
  • Location of all cancer types: including all cancer types in the chart without distinguishing it was confusing since it is an aggregate of the other information on the chart

Step 2 – Design: During the design of my #MakeoverMonday viz I tried to fix some of what I felt didn’t work in the original viz. To do this, I tried to keep the visual and colors simple, separate all cancer types survival rate, and show data from all years captured in the source. Below is the original sketch (wire-frame) I created before building out the Tableau viz.

Wire-frame of #MakeoverMonday Viz

Step 3 – Build: Creating the actual Tableau dashboard was the fun part. The most challenging part of this build was creating the spark bar chart originally designed by Adam McCann. Hats off to Adam for figure out the original logic and Tableau wizardry to get the sparklines to end at the tip of the bars. I had to enlist Connor Caldwell for some help to figure out the math when re-creating Adam’s original viz to use a KPI that is an average v. sum.

#MakeoverMonday week 41 Submission: Visualizing 5-year Cancer Survival Rates Through Sparklines


Thanks for reading!

Please like, share or comment and let me know your feedback and thoughts. 

Who else is drinking a morning coffee?

Ever wonder which countries consume the most coffee? Or if you drink more coffee than the average Joe in your region? I’ve created a Viz to understand just that.

This is my first blog post on where I will share beautiful and fun data visualizations. If you know me, you know I love coffee, so I figured no better way to start this blog than attempt to visualize data about coffee through latte art. Check out my first Viz below.

Click here OR the image above to view the Tableau workbook

The Data: I pulled together data from USDA coffee consumption data  (2017) and blended it with  Wikipedia UN population data (2017).

Design Techniques: A colleague Meera Patel encouraged me to make the design a bit more trendy by using a broken grid layout and weaving in some Serif font, both concepts from this excellent article on “19 web design trends for 2018” by John Moore Williams. You can also check out the color palette I created here.

Tableau References: The most challenging component of this Viz was the sunburst diagram I used to create “latte art.” Shoutout to Bora Beran for pioneering the Viz and Super Data Science for the video tutorial and reference workbook to help build it. And shout out to Connor Caldwell for helping tidy up some of the Tableau actions and formatting.

Thanks for reading!

Please like, share or comment and let me know your feedback and thoughts.