I have data that are in fact percentages of C (bound to clay) combusted in a certain temperature interval.

Exemplary data:

    C300-600 C300-450 C300-900
1       0.96     0.72     0.99
2       0.96     0.70     1.00
3       0.95     0.67     0.99
4       0.95     0.66     0.99
5       0.95     0.67     1.00

Is it meaningful (given the ecological hypothesis is reasonable) to represent the percentage values in boxplots? I.e., showing the median, quartiles etc. of percentage values?

  • $\begingroup$ It depends how you generated these percentages. But in general it is possible yes. Boxplot is just a way to show the distribution of your data in a non-parametric way. $\endgroup$ – peteR Jan 19 at 17:41

There's no reason why not, but other plots may be better, depending on how much data you have and what you want to show.

With N < 50 or so per group, you can use a strip plot and show all your data (with a box overlaid, if you like).

There are also variations on the box plot that display more information.


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.