« Adulterated education | Main | Gauging the water level »


Hadley Wickham

Mosaic plots will still work - they'll be dividing up something other than 100% but that's ok.


I think it's a nice touch that they align the final, incomplete row of filled squares according to the neighbour that is higher where possible, giving a smoother profile.


Nice post! But I'm wondering if one can do a better job ordering the activities in your profile chart? For example, I'd put Collectors just before Inactives, and Joiners in the beginning. Moreover, I'd use hue/saturation among the greens and blues to help trace back the exact age category.

Another chart of interest:


Wouldn't the problems be largely solved if, in the original chart, each "block" stood for a certain number of people rather than a certain column percentage? That way the relative distribution could be read easily across either columns or rows.

This is a common problem when percentages are used in tables as well: are the percentages of the columns? of the rows? of the total? A sloppy table won't tell you.


Aleks: I like both suggestions.

zbicyclist: Using raw numbers sometimes help. In this case, the raw numbers represent the number of survey respondents, which is not useful. For surveys, the only relevant data are the percentages.

Fun Dates

Both are pretty, but why "Merrimeckos"/mosaic charts don't work?

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