My question is rather simple but it seems that statisticians have different views about it.. If a bar chart shows the percentage of children under 5 that had fever in the past 2 weeks by country.. The title will be: "Children under 5 with fever in the past 2 weeks, by country, 2017" The x axis title will be the name of the countries The y axis title will be (for me): "% of children under 5"..as it is the denominator! But some people will write: "% of febrile children under 5" (which for me is not the title of the axis going from 0 to 100...the colored bar will be febrile children, the rest of the bar which is not colored are children who did not have fever...)

What is the correct standard way of labeling the y axis?


This is really more a question of English usage than statistics, but I'd say neither of the choices that you list is optimal. If I had to pick from those two, I'd pick 2, but I prefer:

% of children under 5 who are febrile.


I think your mindset about this is correct. In my opinion, the title should tell the reader what this graph is all about (sometimes a subtitle could help reduce a long title into two parts).

The axis should be quick and easily understood by the reader, but I think the two y-axis titles you mentioned apply to two different representations:

  1. "% of children under 5" - This is when your graph is showing two different populations, namely children who had fever, and children who didn't have fever, and each group is represented with a different colour (and specified in the legend). This would be two bars for each country (positioned to your liking).
  2. "% of febrile children under 5" - This is when your graph is representing one population, namely children who had fever. This would be one bar for each country.

Personally, I think option 2 here is the better one because:

  • It links directly to your title which is "Children under 5 with fever in the past 2 weeks, by country, 2017", and doesn't mention children without fever.
  • It's simpler; if you're showing one bar that's showing a percentage, say 20% for country X, then you automatically know 80% of children didn't have a fever. If you had to represent more than two groups or if your two groups didn't represent the whole population, then Option 1 would have been better!

Hope this has helped!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.