1

I would like to visualise ordinal data with ranks (1st-4th) in different categories however, I'm unsure how to best go about it. I'd like to be able to see the distribution and the mean ranking for each group also as the graph is supposed to add a visualisation to a Friedman test. The graph will be presented in a powerpoint at a conference and must therefore be easy to understand. I've previously used violin plots, however as far as I understand that's more appropriate for continuous data. I've also tried using a dot plot but since the data is ranked, it just gives me four straight lines of dots.

Any advice would be greatly appreciated.

Here is some mock data to work with

The real dataset is much larger with 1248 observations (78 participants).


set.seed(5931)
dat <- data.frame(subj = rep(1:25, 4),
                  group = rep(c("A", "B", "C", "D"), each = 25),
                  rank = sample(1:4, 100, replace = T))
JuliaM
  • 33
  • 3
  • 1
    Possibly related: [Visualizing Likert Item Response Data](https://stats.stackexchange.com/q/3921/), & [Visualizing Likert responses using R or SPSS](https://stats.stackexchange.com/q/25109/); you just have to think of each likert item as a different group. – gung - Reinstate Monica Aug 09 '19 at 20:47
  • These data aren't literally *ranks*. Did you just mean that the data are ordinal ratings? Also, there is no repeated measures aspect to the data, which might be expected if you want to use Friedman's test. If the situation you really want help with does have repeated measures, can you provide such data (simulated or real)? – gung - Reinstate Monica Aug 09 '19 at 21:14
  • I have edited the question to make it more clear. I've also added a subject column, it is repeated measures. – JuliaM Aug 09 '19 at 22:09

0 Answers0