The Wilcoxon rank-sum test as indicated in your opening slide is not a non-parametric equivalent to a paired-sample t-test. That is reserved for the Wilcoxon signed-rank test. The Wilcoxon rank-sum test is indeed the Mann-Whitney U test (or Wilcoxon-Ma--Whitney or Mann-Whitney-Wilcoxon test).
I thought I was going crazy here. Thanks for re-setting my brain, because that was misleading. I can stop watching the video, because the video is of the Wilcoxon signed rank test but the intro was not.
Question: By excluding Differences equal to 0 do you not make your test more liberal? Or do you adjust your n afterwards? Because it seems to me (and yet I could be mistaken), that there is some trickery going on... ;-) I'll look up my prof's slides, because I can't be doing that kind of stuff if it's semi-legal in the exam -- but thank you for a very reasonably slow video; in which you can marvellously follow the instructions AND have some thoughts on your own! Edit: Actually my prof. too says to throw out any differences that become 0. Indeed it seems to be the norm, I was too suspicios ^^.
The Wilcoxon rank-sum test as indicated in your opening slide is not a non-parametric equivalent to a paired-sample t-test. That is reserved for the Wilcoxon signed-rank test. The Wilcoxon rank-sum test is indeed the Mann-Whitney U test (or Wilcoxon-Ma--Whitney or Mann-Whitney-Wilcoxon test).
I agree with Juan.
I thought I was going crazy here. Thanks for re-setting my brain, because that was misleading. I can stop watching the video, because the video is of the Wilcoxon signed rank test but the intro was not.
Isn't this Wilcoxon SIGNED-Rank Sum Test (and NOT Wilcoxon Rank Sum Test)?
Why did you title the video "Wilcoxon Rank Sum Test" whereas what you did in the video is "Wilcoxon Signed-Rank Test"?
please provide excell command for rank ave for ranking -doing by hand is unreasonable for large N.
Question: By excluding Differences equal to 0 do you not make your test more liberal? Or do you adjust your n afterwards?
Because it seems to me (and yet I could be mistaken), that there is some trickery going on... ;-)
I'll look up my prof's slides, because I can't be doing that kind of stuff if it's semi-legal in the exam -- but thank you for a very reasonably slow video; in which you can marvellously follow the instructions AND have some thoughts on your own!
Edit: Actually my prof. too says to throw out any differences that become 0.
Indeed it seems to be the norm, I was too suspicios ^^.
wrong title for this video !!! this is Wilocoxon Signed Rank
This is not Rank Sum Test, please update your video Title, so someone else doesn't get misguided.