How to tell stories & convey information through visualization

Finishing up the 5th segment of the data-driven journalism online course. Week 5 featured data visualization guru Alberto Cairo (@albertocairo). Some key takeaways from this week that are worth sharing to anybody communicating visually with data.

Visualization is a graphical representation of evidence, a tool for analysis, communication, and understanding. Visualizations are the only way to extract meaning and see patterns and trends in a data set.

Misconceptions in visualizing data

Misconception #1 – Only great designers can create visualizations

Making visualizations depends on thinking visually, not on your computer. Start with a sketch of an idea for visualizing a story.

Misconception #2 – “Infographics is about designing cool pictures”

Creating something attractive will attract readers. Many visualizations are beautiful, functional, but not insightful. Good visualizations offer all three. After attracting readers, you also need to deliver something useful.

The four features that define a great visualization

Functional – The shape of the graphic is adapted to the questions the visualization should help answer.
Beautiful – If it is not attractive, readers won’t stop to read and interact with it.
Insightful – Put your data in context. Insight is the discovery of unexpected or relevant information in any data set. A visualization is created to give readers access and insight to data that they would otherwise not have. Many infographics lack context, meaning that they are not very insightful or relevant. A number on its own is meaningless, it becomes relevant in context.
Enlightening– The information the visualization reveals shapes the perception of the reader.

Rules to keep in mind when choosing the best graphic form:

  • Think about the audience and the publication
  • The visualization needs to be functional but can be different in style
  • Think of the questions your graphic should help answer
  • List the things you think your readers will try to do with the graphic

The goal of visualizing data is to provide access to trends and patterns in the data without having to read the data. A designer needs to anticipate the readers’ questions and facilitate what their needs are.

Discussion Point: Choosing the Best Graphic Forms

Part of this online course includes an open discussion by the participants in each module with the discussion topic provided by the module instructor. The topic of discussion centered around an infographic printed in the New York Times, The Cost of a Colonoscopy Varies Across the Country – The New York Times, with the following discussion points recommended:

  • Is it functional, beautiful, insightful?
  • Is the map the most appropriate graphic form? Does it let you do things with the data, such as comparing figures effectively?
  • If it is not the right way to represent this data, can you think of an alternative?
  • Are there important variables or values that we should include to put this data in context?

asdfa89y9

 

My comments for this discussion:

Frankly, I think this is a disappointing visualization from the New York Times. It’s essentially text on a map – there’s not a lot of visual representation here. It’s not exactly beautiful, but not it’s bad. It’s definitely not very functional for creating accurate comparisons. Your eye has to jump around all over the map to see which values are the highest, then store that to compare to others. It’s only somewhat insightful and enlightening. Adding context and additional information to explore would have made it more insightful, informative, and enlightening. Other variables might include the cost of living index for each city represented to get a more accurate idea of how the costs compare from one city to the next.

The map is not the best graphic form for an accurate comparison which is what I want to do with the data. If the audience is an US audience then we could assume that most readers already know where the cities are located, so the map isn’t that useful. A bar graph would have been better for allowing visual comparisons of the high prices, the low prices, and the range.

I threw a quick bar graph together that, though it may be a little boring-looking, is much more functional and insightful.

Colonoscopy-Prices-Rough

That was incomplete and clunky looking so I finally took some time to create a cleaner, simpler, minimal version:

Colonoscopy-Prices-v2

All in all a very worthwhile course that I would highly recommend.

Finally, here’s a list of recommended articles for the visualization section of the course.

Reading List & Recommended articles: