Wikipedia:独自研究は載せない

Wikipedia is not the place for original research. Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article, and to adhere to what those sources say.

Wikipedia:No original research is one of three content policies. The other two are Wikipedia:Neutral point of view and Wikipedia:Verifiability. The policies are complementary, jointly determining the type and quality of material that is acceptable in the main namespace. They should therefore not be interpreted in isolation from each other, and editors should try to familiarize themselves with all three.

What is original research?

Original research is a term used on Wikipedia to refer to material added to articles by Wikipedia editors that has not been published already by a reputable source. In this context it means unpublished theories, data, statements, concepts, arguments, and ideas; or any new interpretation, analysis, or synthesis of published data, statements, concepts, or arguments that appears to advance a position or, in the words of Wikipedia's co-founder Jimbo Wales, would amount to a "novel narrative or historical interpretation."

Primary and secondary sources

• Primary sources present information or data, such as archeological artifacts; film, video or photographs (but see below); historical documents such as a diary, census, transcript of a public hearing, trial, or interview; tabulated results of surveys or questionnaires, records of laboratory assays or observations; records of field observations.
• Secondary sources present a generalization, analysis, synthesis, interpretation, or evaluation of information or data from other sources.

Original research that creates primary sources is not allowed. However, research that consists of collecting and organizing information from existing primary and/or secondary sources is, of course, strongly encouraged. All articles on Wikipedia should be based on information collected from published primary and secondary sources. This is not "original research"; it is "source-based research", and it is fundamental to writing an encyclopedia.

In some cases, where an article (1) makes descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and (2) makes no analytic, synthetic, interpretive, or evaluative claims, a Wikipedia article may be based entirely on primary sources (examples would include apple pie or current events), but these are exceptions.

Wikipedia articles include material on the basis of verifiability, not truth. That is, we report what other reliable sources have published, whether or not we regard the material as accurate. In order to avoid doing original research, and in order to help improve the quality of Wikipedia articles, it is essential that any primary-source material, as well as any generalization, analysis, synthesis, interpretation, or evaluation of information or data, has been published by a reputable third-party publication (that is, not self-published) that is available to readers either from a website (other than Wikipedia) or through a public library. It is very important to cite sources appropriately, so that readers can find your source and can satisfy themselves that Wikipedia has used the source correctly.

In some cases, there may be controversy or debate over what constitutes a legitimate or reputable authority or source. Where no agreement can be reached about this, the article should provide an account of the controversy and of the different authorities or sources. Such an account also helps ensure the article’s neutral point of view.

What is excluded?

An edit counts as original research if it proposes ideas or arguments. That is, if it does any of the following:

• It introduces a theory or method of solution;
• It introduces original ideas;
• It defines new terms;
• It provides new definitions of pre-existing terms;
• It introduces an argument, without citing a reputable source for that argument, that purports to refute or support another idea, theory, argument, or position;
• It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;
• It introduces or uses neologisms, without attributing the neologism to a reputable source.

The fact that we exclude something does not necessarily mean the material is bad — Wikipedia is simply not the proper venue for it. We would have to turn away even Pulitzer-level journalism and Nobel-level science if its authors tried to publish it first on Wikipedia. If you have an idea that you think should become part of the corpus of knowledge that is Wikipedia, the best approach is to arrange to have your results published in a peer-reviewed journal or reputable news outlet, and then document your work in an appropriately non-partisan manner.

Why do we exclude original research?

The original motivation for the no original research policy was to combat people with personal theories, such as cranks and trolls, who would attempt to use Wikipedia to draw attention to their ideas and to themselves.

However, original research is more than just no personal crank theories. It also excludes editors' personal views, political opinions, their personal analysis or interpretation of published material, as well as any unpublished synthesis of published material, where such a synthesis appears to advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose. That is, any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic of the article. See this example for more details.

Applied to all editors, this policy helps secure our reputation in a number of important ways:

1. It is an obligation of Wikipedia to its readers that the information they read here be reliable and reputable, and so we rely only on credible or reputable published sources. See " What counts as a reputable publication?" and " Reliable sources" for discussions on how to judge whether a source is reliable.
2. Credible sources provide readers with resources they may consult to pursue their own research. After all, there are people who turn to encyclopedias as a first step in research, not as a last step.
3. Relying on citable sources helps clarify what points of view are represented in an article, and thus helps us comply with our NPOV policy.
4. Relying on credible sources also may encourage new contributors. For example, if someone knows of an important source that the article has not drawn on, he or she may feel more confident in adding important material to the article.

Example of a new synthesis of published material serving to advance a position

Editors often make the mistake of thinking that if A is published by a reliable source, and B is published by a reliable source, then A and B can be joined together in an article in order to advance position C. However, this would be an example of a new synthesis of published material serving to advance a position, and as such it would constitute original research. "A and B, therefore C" is acceptable only if a reliable source has published this argument in relation to the topic of the article.

An example from a Wikipedia article (note that the article is about Jones, not about plagiarism in general):

Smith says that Jones committed plagiarism in Jones's Flower-Arranging: The Real Story by copying references from another book. Jones denies this, saying he is guilty only of good scholarly practice because he gave citations for the references he had learned about in the other book.

So far, so good. Now comes the new synthesis of published material:

If Jones's claim that he always consulted the original sources is false, this would be contrary to the practice recommended in the Chicago Manual of Style as well as Harvard's student writing manual, both of which require citation of the source actually consulted. Neither manual calls violations of this rule on citing original sources "plagiarism." Instead, plagiarism is defined as using a source's information, ideas, words, or structure without citing them.

This entire paragraph is original research, because it is the editor's own synthesis of published material serving to advance his definition and opinion of plagiarism and whether Jones commmitted it. The editor is citing good sources about best practice (Chicago Manual of Style and Harvard's student writing manual). In an article about plagiarism, some of the points he makes might be acceptable, so long as he provided links or citations to the sources.

But in an article about Jones, the paragraph is putting forward the editor's opinion that, given a certain definition of plagiarism, Jones did not commit it. Regardless of the fact that his opinion appears to be supported, other things being equal, by the Chicago Manual of Style, it remains the editor's opinion.

For this paragraph to be acceptable in the article about Jones, the editor would have to find a reliable source who had commented on the Smith and Jones dispute and who had himself made the point that: "If Jones's claim that he always consulted the original sources is false, this would be contrary to the practice recommended in the Chicago Manual of Style..." and so on. That is, that precise argument, or combination of material, must have been published by a reliable source in the context of the topic the article is about.

The role of expert editors

"No original research" does not mean that experts on a specific topic cannot contribute to Wikipedia. On the contrary, Wikipedia welcomes experts. We assume, however, that someone is an expert not only because of their personal and direct knowledge of a topic, but also because of their knowledge of published sources on a topic. This policy prohibits expert editors from drawing on their personal and direct knowledge if such knowledge is unverifiable. If an expert editor has published the results of his or her research elsewhere, in a reputable publication, the editor can cite that source while writing in the third person and complying with our NPOV policy. They must cite reliable, third-party publications and may not use their unpublished knowledge, which would be impossible to verify. We hope expert editors will draw on their knowledge of published sources to enrich our articles, bearing in mind that specialists do not occupy a privileged position within Wikipedia.

How to deal with Wikipedia entries about theories

For theories:

1. State the key concepts;
2. State the known and popular ideas and identify general "consensus", making clear which is which, and bearing in mind that extreme-minority theories or views need not be included.

Unstable neologisms, and ideas stemming from one individual who is not an authority, or from a small group of such individuals, should either go to articles for deletion (because they "fail the test of confirmability", not because they are necessarily false), or should be copyedited out.

What counts as a reputable publication?

Reputable publications include peer-reviewed journals, books published by a known academic publishing house or university press, and divisions of a general publisher which have a good reputation for scholarly publications.

For non-academic subjects, it is impossible to pin down a clear definition of "reputable". In general, most of us have a good intuition about the meaning of the word. A magazine or press release self-published by a very extreme political or religious group would often not be regarded as "reputable". For example, Wikipedia would not rely only on an article in the Socialist Workers' Party's newspaper, The Militant, to publish a statement claiming that President Bush is gay. However, if that same claim was in The New York Times, then Wikipedia could refer to the article (and to the sources quoted in the article). The political newspaper could, however, be used as a source of information about the party itself.

Ask yourself some questions when you are evaluating a publication. Is it openly partisan? Does it have a large or very small readership? Is it a vanity publisher? Is it run principally by a single person, or does it have a large, permanent staff? Does it seem to have any system of peer review, or do you get the feeling that it shoots from the hip? If you heard that the publication you are about to use as a source was considering publishing a very negative article about you, would you (a) be terrified because you suspect they are irresponsible and do not fact-check; or (b) feel somewhat reassured because the publication employs several layers of editing staff, fact-checkers, lawyers, an editor-in-chief, and a publisher, and will usually correct its mistakes? If it is (a), do not use it as a source. If it is (b), it is what Wikipedia calls "reputable".

When dispute arises regarding whether a publication is reputable, you can attempt to get more editors involved and work toward a consensus. There is no clear definition, but don't ignore your intuition.

Original images

Pictures have enjoyed a broad exception from the NOR policy. Wikipedia editors have always been encouraged to take photos or draw pictures and upload them, releasing them under the GFDL or another free licence, to illustrate articles. There are several reasons this is welcomed:

• Pictures are generally used for illustration and do not propose unpublished ideas or arguments, the core reason behind the NOR policy.
• Due to copyright law in a number of countries and its relationship to the work of building a free encyclopedia, there are relatively few publicly available images we can take and use. Wikipedia editors' pictures fill a needed role.

A known disadvantage of allowing original photographs to be uploaded is the possibility of editors using photo manipulation to distort the facts or position being illustrated by the photo. Manipulated images should be prominently noted as such and, if they are not, should be posted to Wikipedia:Images for deletion. Even noted as having been manipulated, they should not be used to illustrate articles in the main namespace, although editors are free to make use of them on user pages.

Images that constitute original research in any other way are not allowed, such as a diagram of a hydrogen atom showing extra particles in the nucleus as theorized by the uploader. All uploaded pictures are subject to Wikipedia's other policies and guidelines, notably Wikipedia:Verifiability, and Wikipedia:Neutral point of view.

Related policies and guidelines

Wikipedia:Verifiability

By insisting that only facts, assertions, theories, ideas, claims, opinions, and arguments that have already been published by a reputable publisher may be published in Wikipedia, the no-original-research and verifiability policies reinforce one another.

The threshold for inclusion in Wikipedia is verifiability, not truth.

See Wikipedia:Verifiability for more detailed information, and Wikipedia:Cite sources for examples of citation styles.

Wikipedia:Neutral point of view

The prohibition against original research limits the possibility of an editor presenting his or her own point of view in an article. Moreover, by reinforcing the importance of including verifiable research produced by others, this policy promotes the inclusion of multiple points of view in an article. Consequently, this policy reinforces our neutral point of view policy.

In many cases, there are multiple established views of any given topic. In such cases, no single position, no matter how well researched, is authoritative. It is not the responsibility of any one editor to research all points of view. But when incorporating research into an article, it is important that editors situate the research; that is, provide contextual information about the point of view, indicating how prevalent the position is, and whether it is held by a majority or minority.

Disputes over how established a view is

The inclusion of a view that is held only by a tiny minority may constitute original research because there may be a lack of sufficiently credible, third-party, published sources to back it up.

From a mailing list post by Jimbo Wales, Wikipedia's founder:

• If a viewpoint is in the majority, then it should be easy to substantiate it with reference to commonly accepted reference texts;
• If a viewpoint is held by a significant minority, then it should be easy to name prominent adherents;
• If a viewpoint is held by an extremely small (or vastly limited) minority, it doesn't belong in Wikipedia (except perhaps in some ancillary article) regardless of whether it's true or not; and regardless of whether you can prove it or not.

See Wikipedia:Neutral point of view for more detailed information.

Origin of this policy: the opinion of Wikipedia's founder

Wikipedia's founder, Jimbo Wales, has described original research as follows:

The phrase "original research" originated primarily as a practical means to deal with physics cranks, of which of course there are a number on the Web. The basic concept is as follows: It can be quite difficult for us to make any valid judgment as to whether a particular thing is true or not. It isn't appropriate for us to try to determine whether someone's novel theory of physics is valid; we aren't really equipped to do that. But what we can do is check whether or not it actually has been published in reputable journals or by reputable publishers. So it's quite convenient to avoid judging the credibility of things by simply sticking to things that have been judged credible by people much better equipped to decide. The exact same principle will hold true for history" [1]

Some who completely understand why Wikipedia ought not create novel theories of physics by citing the results of experiments and so on and synthesizing them into something new, may fail to see how the same thing applies to history. [2]

On talk pages and project pages

Like most Wikipedia policies, No original research applies to articles, not to talk pages or project pages, although it is regarded as poor taste to discuss personal theories on talk pages.

A few pages have been created devoted to research into issues related to Wikipedia; for instance Wikipedia:Statistics Department and Wikipedia:WikiProject Wikidemia. These pages may contain original research; that is, research for which there is no reference other than projects in the Wikipedia namespace. Original research that does not have Wikipedia as its object should, however, be avoided on these pages too.

Other options

• Meta-Wiki allows original research, see for instance m:research, m:Wikiresearch, m:Wikimedia Research Network, m:wikiversity, m:category:research, and m:statistics.

• Wikipedia-style websites that allow original research but are not affiliated with the Wikimedia Foundation include Wikinfo, Everything 2 and Urban Dictionary.

See also

• Wikipedia:Neutral point of view
• Google test
• Wikipedia:Verifiability
• Wikipedia:Cite sources
• Wikipedia:Template messages/Disputes
• Inventive step and non-obviousness

References

• Crackpot articles: Mailing list post by Jimbo Wales, July 12, 2003
• "Original research", Jimmy Wales, December 3, 2004
• "Original research" Jimmy Wales, December 6, 2004
• "NPOV and 'new physics'" Jimmy Wales, September 26, 2003
• "NPOV and 'new physics'" Jimmy Wales, September 26, 2003 (followup to above)

Further reading

• Academic Publishing Wiki - a wiki welcoming original research
• Wikiresearch, a proposal for a wiki for original research.