Wikiversity talk:Research guidelines

(Redirected from Wikiversity talk:Research guidelines/En)
Latest comment: 11 years ago by Marshallsumter in topic Ethics

These Research guidelines are for conducting research and posting it on Wikiversity. When approved, these guidelines apply to all the separate projects. Please continue to edit and add to these guidelines, but do so with respect to consensus and the general feeling among editors as to how research should be conducted.

On Wikiversity, we allow:

  • Literature reviews that produce new knowledge (secondary research)
  • Research projects within Wikiversity that use research methods in addition to literature review (primary research)
  • Collaborative research projects involving researchers at institutions outside of Wikiversity

Original research is an investigation involving a hypothesis and gathering of data - that data would then be analysed and the procedure evaluated. Secondary research is an alternative interpretation of the results, or discussion of the analysis - involving new theories but not new data.

These guidelines fall down to two main categories: objectivity and ethics.

See also [[../Archive/En|Archive 1]] - Discussion from November 2006 to March 2007





All research must be ethical. If you are deceiving, harming or using data without consent, that research project should not be posted on Wikiversity. Ethics also includes the preservation of raw data, and not covering inappropriate topics (discuss.)

If you use other people's work, research or data as a basis or starting point for your own research, be sure to verify your sources, as well as adhering to Wikiversity's copyright policies.

  • Should bias here mean, as originally, inclination, tendency, leaning but not necessarily (and hopefully not) blind to other points of view?--Hillgentleman|
  • I think the bias here is where you depart from neutral methods of examining the results. Yes, you have to examine different Points of View when you are suggesting new ideas, but a bias would be where it is in your interest to have a particular outcome.

    For example, a study on male/female intelligence may be biased by a male researcher who would probably want males to come up top. It may not be a conscious effort to deceive or falsify the study - just perhaps giving the in-group (males) the benefit of the doubt a little too often. Most researchers would be able to restrain themselves from being as petty as that, but it is still a bias that should be taken into consideration when you read the study. - Xenon (talk) 07:34, 9 February 2007 (UTC)Reply
  • "Wikiversity cannot permit studies that involve such ethical issues as harm (including distress or embarrassment)..." ⇐ It's nearly impossible to avoid at least some distress or embarrassment, because simply learning that one's assumption or hypothesis is wrong suffices to generate at least some disappointment or chagrin. Research is supposed to get to the ground truth, and sometimes the truth hurts. The appropriate ethical guideline here is to minimize or mitigate any loss of face arising from the unanticipated outcome of an otherwise unpredictable line of research. One mantra in research is to fail, and fail again, but each time to fail better. Thomas Edison tried hundreds of materials in his search for a filament for his incandescent lamp. Finding the solution is oftimes tedious. But the light from the solution shines on forever. Moulton 20:57, 18 September 2008 (UTC)Reply
  • It is true that "distress or embarrassment" are vague and broad. It is better to make it more specific. We need something along this line to protect wikiversity. Perhaps "physical harm", and "emotional distress". Parts of the "embarrassment" issues have already been covered by the foundation privacy policies; minor embarrassment is often unavoidable in any learning process. Hillgentleman| 02:23, 19 September 2008 (UTC)Reply
Emotional states are ephemeral, personal, and largely unobservable in a text-telegraph medium where one cannot see a person's facial expression, attend to their tone of voice, reckon their gross body language. In real life you can see someone blush or hide their face, stutter, blubber, or otherwise wordlessly telegraph their emotional state. In a text-telegraph medium, if they don't candidly report their transient and ephemeral affective emotional state, one is left making wild guesses about another person's mindset, largely unsupported by any reliable evidence of the senses. Over time, one might be able to form a haphazard "theory of mind" about an unseen stranger who is acting erratically, but it's unrealistic to make the faculty of remote blind mind-reading a condition upon which a learning project can be vetoed by a worried party who is merely acting on ungrounded (and undisclosed) fears. At the moment my own personal learning project is to discover the name of the fear of those who are apprehensive about these research projects. Moulton 11:03, 19 September 2008 (UTC)Reply
  • In light of what has been happening at Wikipedia, I find this discussion disturbing. Original research comes in many forms including research on "deceiving, harming or using data without consent" or most disturbing of all "not covering inappropriate topics". Those who may be deciding that these are occurring may themselves be the problem rather than the solution. So too can a reliance on "consensus". If only one voice points out what truly is unethical and is ignored, what is lost is indeed harmful as well as what the majority will do. Just tossing in my two cents. --Marshallsumter (talk) 16:37, 20 November 2012 (UTC)Reply

Talk before change


Suggestions to the research guidelines are very welcome. However, since it is in the heart of character of wikiversity, and it may affect many people in several languages, I would suggest that we talk about any change beyond cosmetics before we actually write it down on the actual page. Hillgentleman| 22:41, 18 September 2008 (UTC)Reply

I await comments on the #Ethics section, above. Moulton 23:36, 18 September 2008 (UTC)Reply
I propose that we try to stimulate a "second round" of development of research policy for Wikiversity. First, we need to make a new effort to find additional collaborators who have an interest in research and who have time to help guide the development of good research practices at Wikiversity....this means posting calls for participation in discussion forums and places like Wikiversity:Announcements. I'd particularly like to find a way to involve non-English speaking Wikiversity participants. The main reason for a "new push" on research policy development is the need for community discussion of the Ethical Management of the English Language Wikipedia research project. I am still learning about action research, but the claim has been made that the "Ethical Management of the English Language Wikipedia" project is an example of "action research", so it might be useful to make sure that "action research" is explicitly covered by Wikiversity research policy. We need to address this: "One idea that I would like to propose is an explicit ban on "case studies" using real examples of non-notable people, in exchange for hypotheticals". --JWSchmidt 16:03, 21 January 2009 (UTC)Reply
I think noteworthy alone is too vage to be a useful criteria. I think a section on case studies would be more appropriate. Case studies don't necessarly mean studying people. There is group dynamics or statistical purposes in doing case studies as well. Like how many dogs are put down each year in Tornato, or how many people in Southern California love blue balloons or red balloons. Case studies involving more personal information about the subjects of a case study, should be allowed too if:
  1. People who are the subject of a case study have agreed to be part of the case study,
  2. People's wishes about what information is ok to disclose is respected
  3. People who agree to be part of a case study and to disclosure of some information, but wish not to be identified by name or identifiable, have their information anonymized so as to be unuseful for identifying them.
I think these requirements are more or less in line with what criteria are used for case studies ITRW. --darklama 17:40, 21 January 2009 (UTC)Reply

One of the first things that happened to me on Wikipedia, back in September of 2007, was that I became the subject of an inquiry that went by the official name "Request for Comments". It was manifestly not a scholarly inquiry. I adopted the expression, "Spammish Inquisition" to characterize it. Whatever thinking appertains to this policy discussion here in Wikiversity has relevance to the comparable inquiry processes long in use at Wikipedia. —User:Moulton 19:40, 21 January 2009 (UTC)Reply

"case studies in the real world" <-- I'm not sure of the purpose of making a distinction between the "wiki world" and the "real world". There are many types of research and many kinds of "case studies". We are not concerned with things like medical case studies, rather, we are concerned with the analysis of wiki editing at websites where the edit history is public and it is common community practice to study and learn from that history of edits. I've started exploring some of these issues at Research ethics. --JWSchmidt 03:58, 22 January 2009 (UTC)Reply
"We" should be concerned with clarifying policy on case studies in general. I'm not sure of the purpose in saying "we". I think you might be focusing too narrowly if you are only concerned with problems introduced with attempts to anaylize edits at public wikis. My only purpose in referring to the real world is to point out that Wikiversity could benefit from what works and doesn't work in the real world, rather than trying to reinvent the wheel or pretending like Wikiversity exists in a vacuum. Maybe there are laws which would effect what people must do when doing research that Wikiversity needs to consider? Maybe there are certain procedures that Wikiversity participants need to follow to ensure they are following any ethical requirements required by law? Maybe there are certain procedures that Wikiversity participants should follow to ensure that they aren't violating Wikimedia:Privacy policy or any other WMF policies? Maybe there are procedures that Wikiversity participants should follow in order to do no harm? Maybe there are procedures that Wikiversity participants need to follow just in order to gain people's trust and willingness to be the subject of a Wikiversity case study? Wikiversity could benefit from borrowing ideas from reseach and ethical requirements that different professionals are expected to follow, including medical professionals (Research Ethics). I think without people's trust and confidence in Wikiversity, Wikiversity has little hope of successfully being able to pull off any research or case studies. --darklama 05:28, 22 January 2009 (UTC)Reply
The Wikimedia:Privacy policy applies to authorized agents of WMF who have access to non-public registration records. That policy is silent on what pedestrian editors can do with public information disclosed on- or off-wiki by editors and their occasional correspondents.

Regarding conventional policies in academia regarding ethical considerations when studying human subjects, those are widely employed throughout academia, and many exemplary policies can be readily found.

Whatever thinking goes into a policy in the context contemplated here also applies to conventional RfC's and RfAr's where the practices of editors are called into question and examined for propriety.

Barry Kort 13:44, 22 January 2009 (UTC)
I think you might be focusing too narrowly <-- The existing Wikiversity research policy was designed to cover many types of research, including many types of case studies, by placing limits on research that would normally be formally reviewed by an Institutional Review Board. For example, if a wiki editor decided to create a medical case study then that would be very ambitious for Wikiversity. Patients have an expectation that their medical histories will be kept private. A project that would involve publishing at Wikiversity a person's medical history and health records would have to be done after approval of the research proposal by an Institutional Review Board. We could start making an explicit list of all the types of case studies that require IRB approval, but what I think we should focus on now is the type of case studies that naturally arise in the specific kind of wiki-centric research project that some wiki editors have proposed to forbid from taking place at Wikiversity. I've started a page section for exploring issues specifically related to this particular type of case study. --JWSchmidt 15:02, 22 January 2009 (UTC)Reply

I think you are misunderstanding me. I think Wikiversity could benefit from expanding on the ethical requirements for research rather than adding an explicit list of case study types. I think most types of case studies have similar ethical requirements regardless of whether its mediacal research or some other type of research. Wikiversity could even benefit from Wikinews' Code of Ethics despite it having not been formally adopted yet. Journalists often need to do research for a story. Unwelcome research only begs the question what kind of research is left that people can do? --darklama 16:17, 22 January 2009 (UTC)Reply

Recent edits


User:AFriedman has made some changes [1] to the policy. There are simple improvements on the current text, and also some changes in the content of the guidelines. I have reverted them for the moment. What do people think about them? <Hillgentleman| ~ | > 00:28, 15 February 2009 (UTC)Reply

I think guideline and policy pages should be living documents, open to being edited. --JWSchmidt 15:56, 15 February 2009 (UTC)Reply
Yes and no. Policy and guidelines affect many people, and are results of many and reflect the current consensus. Substantial changes require discussion.
And what about the edit themselves? I don't think the result is better than the older version. In particular, "reuqires the permission of a custodian" doesn't make good sense, because custodians are not vested any authority by the community to make any decision; and decision on wikiversity should be by the community consensus. <Hillgentleman| ~ | > 02:57, 16 February 2009 (UTC)Reply
I was invited by another user to work on the research guidelines, after building a research page on WV and participating in discussions about the research policy over there. I'm pretty new here. What do custodians do? Also, what are the policies about closing pages such as the Wikiversity main page to new edits? I was not trying to change a policy when I brought up custodians, but rather explain that some pages (e.g. the main page of Wikiversity) may be closed to edits but the rest are open to edits. --AFriedman 04:04, 19 February 2009 (UTC)Reply
Thank you for your contributions. The page is open to edit; and edits which are obvious improvements will stay. However, since this page affects all wikiversities, we ought to be careful. A wikiversity custodian, or sysop, is someone who the community trust enough to give them rights to protect, delete and block - that is all. Country Mike on English wikiversity, who also works on Eric Moller's wikieducator, wants custodians to be viewed as representatives who can be liasons to other educational wikis. This is an interesting idea, but it hasn't gained momentum.
Wikiversity tried to keep its front page open for a long time. In general, pages are open to edit unless there is a serious risk of disruption. Some study groups in wikiversity may wish to keep their working area closed to themselves for the duration of their collaboration. This is explained in en:wikiversity:page protection templates. <Hillgentleman| ~ | > 04:19, 19 February 2009 (UTC)Reply
How does Wikiversity benefit from and how is education, learning and research improved by, supporting study groups that wish to keep their working area closed? That's one of the things that continues to puzzles me about Wikiversity. I think Wikiversity and students/learners would benefit and learn more from having everyone whose interested in participating doing so, rather than just a fraction of the people interested in doing so participating.
Basically I see "people have many reasons for wanting this", but no specific reasons mentioned, no examination of the reasons done, and no explanation of how this fits in or supports Wikiversity's mission. Surely without a careful examination of why this is a good thing™ before allowing it, Wikiversity risks creating a closed environment where people see no need or reasons to collaborate, contributors cannot educate, learn or do research without starting a new page, and learners/students cannot learn without reading a large number of pages? <rant/> --darklama 17:23, 23 February 2009 (UTC)Reply
Let me reply only to the first question, for the rest rely on assumptions on what answer is. I said "for the duration of their collaboration". The work can be reused by anybody after they have published their results. There are experiments which could use wikiversity for collaboration, and in these researchs you wouldn't want just anyone, but those who know what they are doing, to edit (some of ) their pages (e.g. the reports and the data). <Hillgentleman| ~ | > 06:31, 25 February 2009 (UTC)Reply
After they publish their results where? Since resources could theoretically continue be developed indefinitely, how would one determine what a collaboration duration is or when a collaboration process has finished? Wouldn't the average person who wants to contribute to a research project either know or think they know what they are doing? What about other study groups doing similar research? Couldn't letting other study groups doing similar research help reinforce the quality of research? Couldn't there contribution be used as a form or peer review of the reports and data? Why should one person or group be able to decide who is qualified or unqualified to help in the research? Couldn't someone even if they turn out not to know what they are talking about, help through there attempts to contribute aid in improving the quality of the research, reports or data. I think study groups can benefit from having fresh eyes taking a look at there research, reports or data. --darklama 02:06, 26 February 2009 (UTC)Reply
New ideas are often very fragile and opposed by the majority. An important part of learning about the world and exercising academic freedom is the process by which scholars are free to select their collaborators and explore unusual or unpopular ideas. In a wiki, if someone creates a resource and restricts who can edit that resource, others are free to make a copy and edit the resource as they see fit. "Wikiversity risks creating a closed environment where people see no need or reasons to collaborate" <-- I see no evidence to support this claim. We are talking about a "safety valve" that is used to restrict collaboration when there is a good reason to do so. The existing research policy depends on the idea of scholarly ethics and a Review Board, both of which function to assure that research scholars remain engaged with the community and work in harmony with the project's mission. These topics were long discussed by Wikiversity participants until a gang of abusive admins took over and started imposing bans and blocks on scholars and imposing their absurd brand of censorship. I totally reject the Wikipedia model of "one page fits all" that is advocated (above) by Darklama. Wikiversity was founded on the idea of being open to doing things differently than how they are done at other wiki projects. It is destructive to the mission of Wikiversity for a gang of bullies to take over the Wikiversity project and try to impose the rules of Wikipedia and Wikibooks on this project. --JWSchmidt 14:38, 25 February 2009 (UTC)Reply
Are you saying that one good reason people need to be able to restrict who can edit pages is because there views might not be mainstream? Wikiversity would hardly be the first Wikimedia project to accept such works. I don't think such works need that kind of protection. Resources can have a defined scope or perspective that needs to be adhered to, and anyone who agrees to adhere to a project's scope or perspective should be free to contribute to the project. I don't see what benefit there is to learners and students to have multiple copies of a work in which the only difference is who was allowed to contribute to it. If group A wants to write about Intelligent Design and group B also wants to write about Intelligent Design, why should group B have to fork group A work's rather than help to contribute to the same work? What "safety valve" are you talking about? My questions, comments and concerns aren't with the research policies or review board, but with en:wikiversity:page protection templates. How does en:wikiversity:page protection templates function to assure that research scholars remain engaged with the community and work in harmony with the project's mission? I don't think it does.
I am not advocating any Wikipedia model, and your assumption shows how little you know about my wiki activity or background. I am also not advocating a "one page fits all" model either. The purpose of my comments is not even about my opinions or suggested model, but about the need for people to clarify whatever the purpose of en:wikiversity:page protection templates is, when it is and isn't a good reason to use page protection, how long its to last, or whatever. --darklama 02:06, 26 February 2009 (UTC)Reply
  • Darklama, you may recall that the Wikversity course on Applied Ethics was open, with the proviso that all participants adhere to the Wikiversity Policy on Scholarly Ethics. Notwithstanding that good faith open invitation, a small but vocal minority (no more than a half-dozen non-resident editors arriving at Wikiversity from Wikipedia) departed from Scholarly Ethics by vandalizing and shredding the work of the scholars who had crafted the course on Applied Ethics. In view of the furore that followed, it's clear that neither open nor closed project design would have saved it from the savage raid that utterly destroyed the course on Applied Ethics. —Moulton 16:04, 25 February 2009 (UTC)Reply
    I am aware neither approached would of worked to deal with that particular problem, nor do I think it was or is intended to. --darklama 02:06, 26 February 2009 (UTC)Reply
  • Exactly as John said. Wikiversity encourages exploring specific points of views as long as they are disclosed. Forking is encouraged. Multiple pages on multiple points of view are fine. In real-life research, a bigger group is not necessarily more efficient. And if you are in a research group, you would really want to make sure only the right people touch your data. <Hillgentleman| ~ | > 08:41, 26 February 2009 (UTC)Reply
    I'm not talking about multiple pages on multiple points of views. Each POV should have its own page. I'm talking about forking of pages on similar points of view or similar research. In real-life similar research studies are done all the time, which can strength a theory or poke holes in it, and functions as a form of peer-review. A bigger group might be able to generate more data. With your repeated mentioning of data, I assume you believe that research data is what needs protecting? Perhaps that proposal needs to be specific about that? Maybe "Research data should only be changed by members of the research group. You are however free to analysis and contribute to the theory or hypothesis surrounding the research data"? --darklama 12:55, 26 February 2009 (UTC)Reply
  • Research data and case studies should not be tampered with by adversarial editors. Rather, adversarial researchers should provide their independent analyses and criticism of research data and case studies, and provide further research data and additional case studies of their own. In a Learning Organization, where the subject of a Case Study is invited into the process of helping to document and analyze their own case study, that is called Action Research. It would be quite appropriate to employ Action Research in a project like Wikiversity, when the subject of research is the efficacy of the dynamics of a Learning Organization. —Moulton 14:59, 26 February 2009 (UTC)Reply
    Anyone should be able to contribute to a case study regardless of whether they are agreeable or not with the conclusions. People should be able to remove conclusions not supported by the available research data, add contradictory conclusions if supported by the available research data, and add an explanation as to why a drawn conclusion is incorrect based on available research data. Any theories, hypothesis or conclusions that could be true based on available research data should be available on the same page for people to read, otherwise people will only see part of the picture rather than the whole picture. Any theories, hypothesis or conclusions that cannot be adequately drawn and backed up based on available research data should be removed until such time as new research data is available that does support it. If a page says "Y is X" than that page should include any supporting research data that proves "Y is X", and any supporting case for why "Y isn't X" if someone feels that can also be drawn from the available data in order to make the theory, hypothesis or conclusion as complete as possible. If why "Y is X" is explained across several pages than the cases for why "Y isn't X" on each page should be in terms of what is specifically covered on each page. --darklama 14:51, 27 February 2009 (UTC)Reply
  • If a critic believes a research conclusion is not supported, they are welcome to present evidence, analysis, and reasoning that falsifies the hypothesis in question. It is not appropriate to summarily delete the falsified thesis; rather the falsified thesis needs to be openly demonstrated as unsupported by evidence, analysis, or reasoning, or otherwise overthrown by new information supplied by the critic (and subjected, in turn, to scholarly peer review). —Moulton 16:04, 27 February 2009 (UTC)Reply

Quite an interesting discussion. I would think that when people are trying to create multiple pages on similar topics, it's not always clear to third parties whether these pages represent distinct points of view, or whether the people who created the newer page were simply unaware of the older page. I have seen people making pages that are similar to pages that already exist, and what I have done is let these users know about the older pages. Then they can explain which one it is. --AFriedman 23:26, 28 February 2009 (UTC)Reply

Experimental Politics


When the Custodians act unilaterally to impose fiat, are they engaging in live experimental research to find out what kind of Liminal Social Drama emerges from their reprisal of anachronistic governance practices that the rest of the civilized world pitched on the rubbish heap ages ago? —Moulton 17:59, 16 February 2009 (UTC)Reply

Peer review from external entities


Regarding the section on peer review from external entities, I'd like us to reach a consensus regarding what quality level is required on the various steps of in order to justify the inclusion of a previously unpublished work in Wikiversity. Such quality assurance measures may include, for example, credentials of the author of the work (such as academic degrees and/or previous experience across Wikimedia projects), credentials of the peer reviewing entity, state of the peer review certification, and/or the importance of the subject at hand (such as an existing Wikipedia article on the matter).

In my opinion, I think a previously unpublished scientific work that has undergone peer review verification can be included in Wikiversity if the peer reviewing entity is trustworthy and the following are stated in the peer review certification:

  • A disclosure of conflicts of interests between the peer reviewing entity and the author of any relations that could affect the issuing of the peer review certification.
  • A statement that the peer reviewing entity considers the method and interpretation of the results to properly support the conclusion of the study.

Any other ideas? Mikael Häggström (talk) 11:22, 9 November 2012 (UTC)Reply

See also


en:Wikiversity:Peer review verification

Return to the project page "Research guidelines".