Round things, square things
Is this chart rotten?

Story within story, bar within bar

This Wall Street Journal offering caught my eye.


It's the unusual way of displaying proportions.

Your first impression is to interpret the graphic as a bar chart. But it really is a bar within a bar: the crux of the matter - gender balance - is embedded in individual bars.

Instead of pie charts or stacked bar charts, we see  stacked columns within each bar.

I see what the designer is attempting to accomplish. The first message is the sharp decline in gender equality at higher job titles. The next message is the sharp drop in the frequency of higher job titles.

This chart is a variant of the "Marimekko" chart (beloved by management consultants), also called the mosaic chart. The only difference being how the distribution of jobs in the work force is coded.

The Marimekko is easier to understand:


A key advantage of this version is to be found in the thin columns.

Here is another way to visualize this data, drawing attention to the gender gap.


In the other versions, the reader must do subtractions to figure out the size of the gaps.


Feed You can follow this conversation by subscribing to the comment feed for this post.


In the last image, what is the point of plotting both men and women information? 4% more men is the same as 4% less women. And what is the gap then, is it 4% or 8%? If you want to plot the gap just plot the gap.


I prefer the Marimekko, but wonder if the information about the number in each group is really necessary. For a scientific publication we could just display the proportion for each group and a 95% confidence interval.


A: I'm pretty sure I have made that comment myself somewhere in the archives. But there is always a balance between efficiency and "entertainment". In this case, I opted to show the gap as a literal gap between two lines, feeling that it makes for a more interesting chart, without making it unreadable.

Ken: The marimekko has the advantage that the total area represents 100% of the population but whether that is important to show depends on the researcher's message.


Kaiser: I thought it wasn't important here. What would concern me is whether the proportion is reliable in the small groups, and so confidence intervals are more useful.

The comments to this entry are closed.