The big difference between David McCandless and Hans Rosling

*My opinion follows, treat it as such*

If you have any interest in infographics and data visualization you probably know the names David McCandless and Hans Rosling.  They are both data viz superstars but there is a big difference between the two.  That difference is their approach.

The David McCandless Approach
1. Have an argument that you would like to make
2. Find data that could support your argument
3. Visualize and present the data

The Hans Rosling Approach
1. Have an argument that you would like to make that is based on solid evidence
2. Supporting data is already known
3. Visualize and present the data

For McCandless, the data is window dressing for an argument he developed even before finding the evidence.  For Rosling, the data is evidence that underpins the argument.  McCandless uses visualization to help you understand his argument.  Rosling uses visualization to help you understand the data, which strengthens his argument.

I don’t believe I am splitting hairs.  McCandless approaches data as a journalist and Rosling as an academic.  It’s the reason why Rosling has one big presentation he tweaks and repeats while McCandless has a book of visualizations.  Understanding the data is a lot of work and, with constant content requirements, it’s not always feasible as a journalist.  It’s one of the major reasons that journalists often turn to experts.  Rosling, on the other hand, is an expert.

There is nothing wrong with McCandless’ approach, it is what it is, but it’s important to understand that it is different.  Malcom Gladwell is not an academic but don’t let that stop you from reading The Tipping Point.  Is a McCandless visualization based on data? Yes.  Is the data accurate? Sometimes.  Is it accurately depicted? Sometimes.  I’ve critiqued his work in the past (you can read the post here) and I won’t rehash my argument.  I’ll just say that the journalistic approach to data visualization often misses important context or fails to consider the reliability of the appropriated dataset.

As media organizations and bloggers become increasingly data savvy the line between academic and journalist gets fuzzy.  Seeing data visualizations in newspapers and on blogs is great, but always remember to question the data and view the viz with a critical eye.  Even if the one presenting is a superstar.


  1. vizsceptic on May 17, 2013 at 6:03 am


    I would go further, though. Good journalism separates clearly the reported facts from the inferences and comment. McCandless mixes the two dishonestly so that nobody can see the join and presents his angle as fact.

  2. Chris Lysy on May 17, 2013 at 8:10 am

    Thanks for the comment. I agree, his work isn’t a prime example of good journalism, or at least it wasn’t at the time I wrote this piece. I haven’t been following him much as of late.

    I wrote this piece two and half years ago after having seen some really bad data visualizations. I started rethinking my previous position on McCandless because creative work is tough and at least he presents his sources. I know that’s a low threshold but so many visuals failed to meet the same test.