« A challenge | Main | Points of comparison »


Jorge Camoes

Probably there are circumstances where you can use this type of chart, but not in this case.

A simple bar chart with overlapping series should be enough, and it is not at all intimidating. I've been playing with overlapping series in population pyramids and not only you have a clearer view of proportions by age group but you also maintain the "pyramid" profile.


I think the scatter chart is best. I had hoped for some nice insights from my favourite scatter chart trick, the grid of diagonal lines, but nothing much appears to jump out.

It appears that the largest advertisers had the lowest proportional unmeasured spend while the smallest (among the majors) had the highest. (It's only a weak correlation: a linear fit yields only 16% R-squared.)

If instead you hypothesise that big advertisers spend a quarter as much on unmeasured media as on measured media, plus another half billion or so, the R^2 is 43%. y=0.254x+480 [chart]

The half billion accounts for the increase in proportion among smaller advertisers. I don't say 43% is good, only that it's an indication of how comparatively weak an inverse linear relation between total ad spend and percentage unmeasured media is.


Although the scatter plot is better for analytical purposes I think that it would be above the heads of 99% of NYT readers. I guess this kind of consideration really limits the kind of visualization that could be considered.

The comments to this entry are closed.


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
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