Wilcoxon Rank Sum Test vs Van de Waerden Test

#1
Hi,

Would anyone be able to explain the difference between the Wilcoxon Rank Sum Test vs Van de Waerden Test? Do they look at or analyze the same thing? Can they be used interchangeably? Why would you use a Van de waerden test over a Wilcoxon rank sum test? What advantage or disadvantage does the Van de waerden test have over the Wilcoxon rank sum test? When would it be appropriate to use the Van de waerden test over the Wilcoxon rank sum test and vice versa.

Any help would be greatly appreciated. By the way I'm not a statistician and I only know very basic statistics, so... if there's a way to explain these in the most elementary way, that would be awesome.

Thanks!
 

trinker

ggplot2orBust
#2
This may get you started.

I myself am not familiar with the Van de Waerden Test and so your query challenged me to learn a bit about it. As far as I can tell is that the 2 are not used interchangeably. The Wilcoxon Rank Sum Test works off the logistic distribution whereas the Van de Waerden Test works off the normal distribution. This off course is scratching the surface. Probably your best bet to gain better understanding is to Google it, as I did, and start parsing through different sources.

Thanks for the question, it caused me to look at something new. :)
 
Last edited:

Dason

Ambassador to the humans
#3
Congrats trinker on getting TS Contributor status! This was the first time I noticed the change in your title.
 
#7
Hello

I am new in this forum.

Both Wilkonson and Van der Varden tests are nonparametric rank tests for comparing to two data sets
The difference is Van der Varden test is applied only when you know that two data sets have equal standard deviations.

And for the same number of sample size Van der Varden test has less second kind of error than the Wilconson test has