# How to get the standardized z-statistic from dunn.test-package in R?

Sorry if this is a trivial question, but I couldn't find anything on it.

I am running a Dunn-test with the dunn.test-package:

dunn.test(data$climatescepticism, g = data$mediaoutlet, method = "bonferroni")

I get the following output:


data: x and group
Kruskal-Wallis chi-squared = 28.2413, df = 3, p-value = 0

Comparison of x by group
(Bonferroni)
Col Mean-|
Row Mean |          1          2          3
---------+---------------------------------
2 |   4.638266
|    0.0000*
|
3 |   0.166923  -4.445118
|     1.0000    0.0000*
|
4 |   2.842656  -2.566897   2.592805
|    0.0134*     0.0308     0.0286


To report this results in my thesis I need the standardized test statistic.

Is the Z statistic in this output already standardized? And if not, how is it possible to obtain the standardized value or at least get the standard error to calculate it myself?

• "Sorry if this is a trivial question, but I couldn't find anything on it." Perhaps try e-mailing the author at the supplied address in the documentation? Commented Dec 18, 2021 at 19:23

## 2 Answers

The $$z$$ test statistics are reported in the output of dunn.test, as you can see in your supplied output (e.g., 4.638266 is the $$z$$ statistic, and the starred 0.0000 beneath it is the $$p$$ value). Aside: all $$z$$ statistics are ‘standardized’ in that they derive from the standard normal distribution.

If your use the list=TRUE option, you will get the same numbers ($$z$$ statistics and $$p$$ values) arranged in list form also.

if you want to get the z-score, try installing the package "FSA" instead. See the image of the code below. Cheers.

• –1 dunnTest in the FSA package is just a "rewrapping" of my dunn.test package and offers no additional functionality. Commented Dec 18, 2021 at 19:20