Flower People

My latest visualization is a different aesthetic take on Gapminder data. Specifically this looks at the relationship between the change in life expectancy and income from 1965 to 2015 by country and region. This viz was inspired by Valentina D'Efilippo's amazing Poppy Field visualization. Her work is awesome. Check it out here.

The size of the flower represents current population. The height of the flower represents GDP per capita. And the curvature of the stem represents the change in life expectancy between 1965 and 2015. Click the image to interact.

I also think this visualization is interesting in light of the conversation/debate on Stephen Few's blog regarding the debate between data visualization as an art or science. Few is firmly in the latter camp with others like David McCandless in the former. In my humble opinion there is a time or place for both. This visualization falls more in the art camp obviously. Is it harder to read than perhaps a traditional chart but visually it's more compelling. 

While I don't think a visualization should completely forsake accuracy for art I also think telling a compelling story is part of visualizing data. Ideally, your data visualization should invoke thought or emotion in an honest manner. Now some visual mediums are better for conveying data accurately. And some are more visually compelling and inspiring. Finding that nexus is the fun part of this field.


  1. I think if the paths represented the actual values through time instead of a simple spline (or bezier) you'd have something that is both artful and accurate all the way through.

  2. Thanks for your comment Robert. I don't think your approach would be more accurate but more detailed. Also it wouldn't be very artful as the lines wouldn't be smooth but erratic due to events (war, poverty, etc). They would look like dying flowers.

  3. Maybe I should explain what I mean by "accurate" since that may not be the best word to describe what I mean. When I see lines connecting data points, I'm accustomed to thinking the points along the way represent data, especially if it's not a straight line between the points. So, I see a point on the line for Oman corresponding to 57-58 years at $20K GDP Per Capita. But, I know after looking at it a little bit that's not the intent of the line. It's there for design purposes other than communicating details of the data.

    It's clearly an aesthetically pleasing, attention-grabbing viz that correctly conveys the "up and to the right" feeling and gives a vague notion that some have gained life expectancy without as much gain in GDP. But, the classic "Tale of 100" Tableau viz is also quite interesting to look at while every bend in the line represents an actual data point. Doing something similar with this data would look totally different and I get the design choice to go with a more artful representation.

    Hopefully that explains my point a little more.