You seem to use the 'clusrank' R package. You should read the Article related to this : Article Wilcoxon Rank-Based Tests for Clustered Data with R Package clusrank

or the related PDF in attachment

Qamar Ul Islam , A) Your response doesn't answer the question. B) The information is generally wrong in that neither the Wilcoxon signed-rank test nor the Wilcoxon rank sum test is in general a test of medians. That's the problem with copying information from random websites: they are often wrong. C) It appears your answer is copied directly from other websites. You should probably be aware that a lot of people who use ResearchGate have teaching appointments and we have access to the best plagiarism-detecting software. You might also review the plagiarism policy at your university. I'm sure that they consider copying text from websites and presenting it as your own work is plagiarism. ... As a side note, the first paragraph of your response appears to be copied from here: https://www.statisticshowto.com/wilcoxon-signed-rank-test/ . I find that this statisticshowto site tends to be riddled with wrong or confusing information, as well as having some incomprehensible writing. I wouldn't advise using it.

Qamar Ul Islam , the Wilcoxon is not a test of medians. See, for example,

@article{DivineEA2018,

author = {George W. Divine and H. James Norton and Anna E. Barón and Elizabeth Juarez-Colunga},

title = {The {W}ilcoxon–{M}ann–{W}hitney Procedure Fails as a Test of Medians},

journal = {The American Statistician},

volume = {72},

pages = {278-286},

year = {2018},

doi = {10.1080/00031305.2017.1305291},

}

The Divine et al reference gives some examples that show where a median is higher for one group but the Wilcoxon prefers the other. Here is an example I use where Wilcoxon shows A < C, B < A, and C < B, so intransitive.

A

Sal Mangiafico , good catch. Plagiarizing and not even choosing correct things to plagiarize. I guess forwarding a snap of that to their PhD committee could be useful to warn them that this may be behavior to look for.

Fatemeh Khozaei , what aspect of Qamar Ul Islam do you agree with.

1) That he plagiarises.

2) That what he copied and pasted is wrong.

Daniel Wright, are you using one-tailed tests for the examples you posted above, or have I misunderstood something? I just tried your examples using Stata's -signrank- command, and I'm getting these results for each of the 3 comparisons you highlighted:

z = 1.887

Prob > |z| = 0.0592

Exact Prob = 0.0585

Lower one-sided exact p-value = .97072605

Upper one-sided exact p-value = .02927395

Full details can be seen in the attached text file. Thanks.

Bruce Weaver , yes, I was using the one-sided in book since that fit into the example, but significance isn't an issue here, it is just the direction of the finding (which is why using the directional test was used, since unlike Stata this is only easily seen from the R output using the one-tailed version). The A, B, and C are different groups in my example, but in general any procedure using ranks can show intransitivity. I couldn't tell from your output if you had these as separate groups (reading off my phone).

Thanks for clarifying, Daniel Wright.

Fatemeh Khozaei – you have a habit of adding "agree with Qamar Ul Islam " as an answer. Be careful. While your answers to other questions appear helpful, his answers are mostly fatuous nonsense, and this one at least is also plagiarised.

As we say, sleep with dogs and wake with fleas.

Update : Qamar Ul Islam 's last paragraph is plagiarised from here: https://data.library.virginia.edu/the-wilcoxon-rank-sum-test/

and his middle paragraph plagiarised from here : https://colors-newyork.com/what-does-a-significance-level-of-5-associated-with-a-t-test-mean/

Ronán Michael Conroy , since the answers after Qamar Ul Islam 's but before Fatemeh Khozaei 's comment showed he was both wrong and plagiarizing, Fatemeh must be pro-plagiarism and/or pro-spreading incorrect information, or it may be something more nefarious like she is just searching for his comments and writing "agree." Let's hope she is just pro-plagiarism and misinformation.

Fatemeh Khozaei wrote the following for an RG response:

Systematic reviews often use statistical techniques to combine data from the examined individual research studies, and use the pooled data to come to new statistical conclusions. However, Not all systematic reviews include meta-analysis, but all meta-analyses are found in systematic reviews.

The following is taken from https://ncu.libguides.com/researchprocess/systematicreviews which reads:

"Systematic reviews often use statistical techniques to combine data from the examined individual research studies, and use the pooled data to come to new statistical conclusions. This is called meta-analysis, and it represents a specialized subset of systematic reviews. Not all systematic reviews include meta-analysis, but all meta-analyses are found in systematic reviews."

I wonder how common plagiarism is in RG answers???

😆 Not everyday you are witness such great teamwork. Great work, fellas.🏆

I think plagiarism is a considerable problem in RG.

I would like to push RG foreward to screen and flag or eliminate plagiarism, but I am not sure how to best initiate this. I unfortunately don't have a list of examples, but such a list would make the severity of the problem more evident. Any ideas?

Daniel Wright – having looked at Fatemeh Khozaei 's contributions to questions, they frequently point to helpful papers or resources. I am willing to believe that when she quoted a very useful piece of information, the failure to attribute it and link to the page (which would have been helpful) was more likely to have been a simple mistake. This is not the only instance of unattributed quotation from her answers. A cursory search rapidly found the sources more of her answers, likewise without links or attribution. And in one case I found that Fatemeh was not the only plagiarist – the text appeared on several websites! So she may plagiarising plagiarists…

It is still not clear to many people that when you use the intellectual work of others you should credit them – it also means that you are making them responsible for their contributions!

However, the whole genre of "I agree with @Professor Gimmeajob" stinks.

What about this : that when you ask a question on RG you become the moderator of the answers. That way the questioner can delete answers that are simply wrong, or plagiarised, or self-promoting or sycophantic. This would leave a cleaner thread in place for reference.

Ronán Michael Conroy Dear Professor, Thank you for your comment. First of all research gate is a platform for exchanging ideas. No one might be perfectly knowledgeable or aware of all aspects of science. We are sharing, learning and growing together. As I might not be agree with some of your answers as well but I respectfully comment. As you know research gate is mostly used by academicians so we should use a proper language in our level. It is not playground for children to fight or using improper words. If you disagree with any of my responses i will hear form you, listen and would be grateful if you also may comment in a proper way. Thank you

I strongly concur with Dr. Fatemeh Khozaei for her valuable comment.

Daniel Wright Thank you for comment. When you answer your students' questions and refer to a website or book and bring the source is it plagiarism? When you surf hours reading about a subject and for showing your student some texts to read is it plagiarism? Are we publishing papers in RG? We are all aware of what is plagiarism. Looking at the Q & A section in research gate as a tool for sharing knowledge and helping others, I believe you need to revise your comments. However I am very grateful for correcting my answer. Here I learned from you and I am so grateful. Keep up the good work.

I advise to Ronán Michael Conroy , Daniel Wright , Sal Mangiafico and all the RG Members. It is not place for debate or criticism. Keep in mind, Kindness makes you a good person so involve in focusing on positive rather than always pointing out offensive comments and use improper words. Be kind and Be motivating here as It shows that there is difference between literacy and illiteracy. Hope it simplifies my point here.

Kind Regards

@Qamar Ul Islam , here is a list of answers of answers you have posted to ResearchGate questions: https://www.researchgate.net/profile/Qamar-Islam-3/answers . Do you think it is not obvious that you cut and paste from websites some general language on the topic that doesn't address the question being asked? Or post some links that also don't contain anything that addresses the question? When you answer 15 or 20 questions per day on a wide range of topics, do you not think it's obvious that you are not adding valuable information, but are just spamming everyone. ... You don't understand what kindness is. Kindness does not involve wasting people's time with spam links and irrelevant plagiarized text. Kindness is taking time to help someone with relevant, thoughtful information, just as others in the past have donated their time and consideration to help you achieve in an academic environment. And, yes, as someone who teaches undergraduate students, I can tell you that it is actually a kindness to point out to someone that it is obvious when they plagiarize material or post meaningless fluff. What reaction do you think this behavior entails in the real world? Do you think employers and colleagues appreciate this? I am sorry that your academic training to this point has failed to provide you this lesson and perspective, that many undergraduates benefit from when they attempt similar behavior.

Qamar Ul Islam , Fatemeh Khozaei

I am honestly curious why you feel the need to copy and paste "answers" from other websites without citation, recommend each other constantly, and write comments that do not address the question in any meaningful way? Since the vast majority of your answers I checked do not match the question, the only reason I can think of is that you attempt to artificially inflate your RG score.

This is why I am curious: Do people you work with actually care about this score? Do you benefit from it in any way? Is it important to you? I could not tell you mine if held at gunpoint.

According to your CV, you are both assistant professors. You must be held to a higher standard as Sal Mangiafico has correctly pointed out. I can only ask you to stop spamming and plagiarising (and reporting your answers, of course).

@Salvatore S. Mangiafico and Marcel Grieger I m not here to fight back and not like a childish field of battle ground like you did. I am normally feel unnecessary to answer back negative people. Hope you understand. I suggest before pointing out mistake thinks of yours first.

I would appreciate you answering Marcel Grieger 's question rather than calling other people childish or negative after you plagiarize and spread misinformation without apology. Several of us having been discussing possible reasons why you and other people do this, and how RG can be improved.

For a general discussion of spam on ResearchGate, see also: https://www.researchgate.net/post/What_can_be_done_about_spam_on_ResearchGate_questions

Good Luck with your criticism debate. I don't wanna be part of it.

Stop Blaming Stop Arguing.

Thank you.

Marcel Grieger If you be aware of academic world RG score doesn't bring you reputation! So first I think you need to rethink and re ask your question. When you would like to apply for job application as I believe you may soon, you will be asked about your number of publications, citations or ability in teaching. Accordingly first you need to fix this perception. Secondly I suggest you to read about the concept of plagiarism for better understanding the concept. You may google it.

Sal Mangiafico I went through many of his answers. I didn't find what so called plagiarism! in many of his responses he has tried to provide a link for the students to read more or even have provided the source of his answers. I am not sure about yours but in our educational community RG score is not a prestigious kind of reward worth spending time to gain more scores! I believe if some one spend his valuable time to help the others and even surf the internet for helping them should be appreciated. I shocked the way you pointed at a researcher and accusing him of plagiarism. By the way I believe here we share our knowledge and through proper criticism we grow. Pointing at one person in harsh manner and accusing him in the way doesn't look proper to me.

Fatemeh, the discussion is about a comment you recommended and commented favorably on, so I think readers will infer that you are well aware of this and engaging in, well, I will let them decide what type of behavior. The message (and yours about it) have been deleted. But, let's pretend you really have no memory of it. Since this is www, here it is.

Your welcome for reminding you about a comment that you were discussing in detail only a week ago. Maybe you really didn't remember it and maybe you have no knowledge of why your previous comment was deleted. I think readers will be able to make their own opinions about your behaviors.

That version may be blurry, here is an easier to read copy. @Fatemeh, I hope this helps remind you of the comment.

Given the distribution of your variable, you cannot apply a one-sample t-test, so you can apply the non-parametric versions that you have mentioned