« Graphical evidence that Bernie got shafted | Main | Well-structured, interactive graphic about newsrooms »



Scatter plot?

Michael Schettler

Thinking about the blocks of age distribution I really question the bins chosen, especially that under 15 bin.

The study was actually a demographic survey of global census data, so I wonder, for each individual census, was the under 15 actually actively involved or was it just the parent, stating the under 15's religious affiliation?

Anywho, I would want to really penetrate the methodology before making any broad conclusions if this is an example of the analytical rigour.


Reminds of a Stephen Few quote that I heartily agree with"

"If you’re inclined to think that a data visualization should limit itself to a single graph, it’s time to leave that constraint behind. The best solutions often require multiple graphs."

(Whether a given piece of software plots these two charts together as one entity or not, this solution is two separate charts, as far as I am concerned)

I also agree with Michael that I find the data itself questionable, considering the focus on under 15, and the dubiousness of their involvement and/or honesty in the process. I am having a hard time finding much useful to actually say from the data, whether plotted well or poorly :)

I don't think I would have set up the colors this way - the connections seem a little weak.

But the one thing I would have done differently for sure is to remove the "Everyone" bar, and instead plot that value as a reference line. I find it considerably easier to comprehend the relationship to a comparison value in that way - each bar is more directly compared to the one line that crosses them all.


I appreciate the more concise view provided by the new dual graph, concentrated color. What I think would clarify the data would be an additional bit of information of family size as the other responses are correct about the viability of the under 15 years old answers. I would offer up that family size choice ( of lack of choice) is a cultural difference with religion playing a significant part. That is something that these two graphs cannot show but might be useful to provide.
Great post, though, very useful. Its just we data folks can never leave well enough alone and want to know the rest of the story.

The comments to this entry are closed.

Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR.

See my Youtube and Flickr.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Keep in Touch

follow me on Twitter