Closed GegznaV closed 4 years ago
This used to be supported in ggstatsplot
, but I decided against it because I wanted to be consistent across functions in terms of how many tests are supported for each type of statistics (always 1). ggcorrmat
was exception where it was confusing whether specifying type == "nonparametric"
was supposed to return Spearman's rho or Kendall's tau.
I think a better solution is to use correlation
package to extract stats details and then use a visualization from a host of packages that support visualization of correlation matrices.
Describe the solution you'd like Add the possibility to chose Kendall's correlation coefficient in
ggcorrmat()
How could we do it? I didn't look how it is implemented in ggstatsplot now, but usually method
"kendall"
is used instead of others incor.test()
.