Wikipedia talk:No original research

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Peacedove.svg The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing. Changes to this page do not immediately change policy anyway, so don't panic.
Frequently Asked Questions (FAQ)
I disagree with the definition of secondary source.
Wikipedia mostly follows the definition in use by historians, which requires more than simply repeating information from some other source or rearranging information from the author's notes. The earliest definition of a secondary source in this policy was in February 2004 "one that analyzes, assimilates, evaluates, interprets, and/or synthesizes primary sources".
WikiProject Spoken Wikipedia
WikiProject icon This page is within the scope of WikiProject Spoken Wikipedia, a collaborative effort to improve the coverage of articles that are spoken on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.


NAC: Whether or not stupid jokes in articles are original research, they are vandalism and are blockable. No need for further discussion. Robert McClenon (talk) 22:02, 27 January 2015 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

What is "original research" in regard to fiction? (talk) 19:30, 27 January 2015 (UTC)

It means we can't add your stupid little joke that you made up about a fictional character in a Japanese children's television series just because you think it's funny and you made up a link to post alongside it. The person on this IP address has repeatedly added a stupid joke sentence to a page on my watchlist [1] [2] [3] and used this non-existant website to support the inclusion. The IP editor is clearly acting in a manner to attempt to game me into violating 3RR, or is acting on the pending arbitration case outcome.—Ryūlóng (琉竜) 20:13, 27 January 2015 (UTC)
Don't you think you're taking that too personally and your pet project too seriously? You have an interest in Japanese culture, fine. But can't you at least change and not take every "joke" edit personally? (talk) 21:13, 27 January 2015 (UTC)
You added a joke to the page. I removed it. This isn't my "pet project". I'm preventing your vandalism from appearing in articles.—Ryūlóng (琉竜) 21:19, 27 January 2015 (UTC)
While I can't deny you being a good contributor, the way to handle others is the main reason of concern. Namely excessively use of profanity, poor communication with others and repeated efforts to get people you dispute with blocked via off-wiki requests though admins and IP checkers you know will fulfill your wishes. Taking the humor edit as a direct attack against you shows you need to work on communicating and not lose your cool. Let that be a lesson to you. (talk) 21:23, 27 January 2015 (UTC)
So yet, again, some random IP editor with a grudge against me has decided to vandalize Wikipedia to spite me. And, please, Mr. IP editor, point out where I was profane to you in regards to the addition of the sentence on the fictional character named "Ticket" or provide proof that I went off-site to ask that you be handled properly, seeing as I posted the page onto WP:RFPP and have been in open discussion on site with another administrator regarding how to handle these articles that you and others have decided to vandalize and disrupt solely to get my goat.—Ryūlóng (琉竜) 21:27, 27 January 2015 (UTC)
The language of your edit summaries speak for themselves. To you is was a "stupid joke". A lot of times you used profanity on your talk page to voice your frustration. You often assumed someone that made edits like mine was blocked or banned and "out to get you". When you reported someone and the block was not enacted, your next stop was SPI. Maybe if you said "that's funny, but it doesn't go here". I also pointed out that fiction is "original research" by nature as anyone can write a story and tell a story. There no way to prove the story or movie plot is true. (talk) 21:35, 27 January 2015 (UTC)
Nobody likes to be told they're wrong, and I'm not surprised to hear your "give me the goat" claim. (talk) 21:47, 27 January 2015 (UTC)
If I'm doing this to spite you, then it means that I'll add the stubhub joke after the protection expires. If you see that again, then you can truly say I'm trying to be spiteful. (talk) 21:53, 27 January 2015 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

"Stupid" is not profanity. And this series of events you describe is untrue. Usually, I reported obvious sockpuppets to WP:AIV because of their current disruption and when administrators do not respond to those requests I have to officially open the WP:SPI case to get the obvious sockpuppetry taken care of in an official manner. I don't see why I can't use my own talk page to express my own frustration at things happening to me with regards to things happening on the project. But this is neither here nor there with regards to your question on writing about fiction on Wikipedia.
Works of fiction are described on Wikipedia as is. It is not original research to describe the events in a work of fiction because the work of fiction is a source for itself in writing about itself. There is nothing against this policy or others that forbids me from writing about King Arthur having received Excalibur from the Lady of the Lake because this is a fictional event and I am describing it in the context of its fictional nature. It is not a violation of this policy to present this information because it is a part of a written work of fiction/oral legend. Now, anything beyond the basic description of the events of a work of fiction requires reliable sourcing. I cannot present my personal literary analysis of Arthurian legend on Wikipedia no more than you are allowed to post what you thought was a funny joke that you made up yourself because that joke does not exist in any form in the original work of fiction. It is not as if there is a running gag present in each episode of that Japanese television series regarding the American company Stubhub, unlike the running gag that the character demands not to be referred to as a handpuppet despite that being the obvious situation to all of the characters. That is something that could be brought up as it is explicitly stated in the source material. Again, anything that is not explicitly stated in a work of fiction requires sourcing otherwise it's original research. I hope this cleared things up for you.—Ryūlóng (琉竜) 22:06, 27 January 2015 (UTC)

No OR is a bad policy, in my view.[edit]

No OR impoverishes Wikipedia. It would be much more valuable if reliable original research could be included. deisenbe (talk) 23:39, 28 January 2015 (UTC)

  • @Deisenbe: I don't agree it "impoverishes" the project. Rather, it provides focus and scope to the content. There could be a choice by the Foundation to at some time provide a section or a new wiki instance for original research, but including that in the encyclopedia proper would change its nature in a fundamental way. I'm not saying that would be good, bad or of no consequence, but it would certainly be a different product. --User:Ceyockey (talk to me) 00:00, 29 January 2015 (UTC)
Are you aware of our sister project called Wikiversity? It was set up in part to give the community a place to present original research. Check it out some time. Blueboar (talk) 00:37, 29 January 2015 (UTC)
@Blueboar: I thought this was primarily focused on housing teaching / tutorial materials rather than being a place for hosting original research. --User:Ceyockey (talk to me) 01:06, 29 January 2015 (UTC)
Wikiversity is mostly about teaching, but Wikibooks: is happy to accept books on just about any subject, so long as it is "instructional". WhatamIdoing (talk) 06:34, 29 January 2015 (UTC)
Back on topic (whether OR is a good or bad policy) imagine if there was no prohibition on OR. Wikipedia is here to allow editors to conglomerate published material into useful articles. No OR policy? Egad, we already have enough trouble herding cats as is. Without the OR policy we would descend into a treacherous blogdom. – S. Rich (talk) 06:58, 29 January 2015 (UTC)
@Deisenbe: the real issue is your word "reliable". Who would decide what is "reliable" OR? You? Me? It just wouldn't work. Peter coxhead (talk) 10:19, 29 January 2015 (UTC)
Of course "reliable" is the issue. "Peer review", done with the reviewers' names usually unknown to the author, though not to the journal editor, is the gold standard. Yet I don't see that WP is incapable of carrying out the function of peer review. Informally, not formally. If you publish OR by an expert, or even by Joe Shmoe, other experts will edit it. If somebody puts in garbage (and there are plenty of published sources for garbage, so insisting on "a published source" is not necessarily infallible), someone else shovels it out, just like now. And to say it has to be "a reliable source" unavoidably involves human judgments, however fallible, about what is reliable.
WP isn't going to change the no-OR rule. I'd be very surprised if it did. But the current system is not without its shortcomings and costs. Just to mention one: if someone with OR comes to WP s/he will be told to publish it elsewhere. At the very least, this brings delay. And those peer review panels are not that infallible. Hell, they gave an academy award to "Driving Miss Daisy" over "Do the Right Thing". And Bush over Gore.
Well that's my 2c for a Thursday morning. deisenbe (talk) 16:43, 29 January 2015 (UTC)
No. They did not give an Academy Award to Bush, and they did give an Academy Award to Gore. Robert McClenon (talk) 22:34, 29 January 2015 (UTC)
Wikipedia is incapable of peer review as practiced by good academic journals. Editors assign papers to reviewers who the editor knows are well-qualified in the subject of the article. Most Wikipedians edit anonymously and there is no mechanism Wikipedians, whether they edit with their real names or not, to establish their credentials in any particular subject. Jc3s5h (talk) 17:00, 29 January 2015 (UTC)
Not to mention we'd be inundated with "research" supporting fringe theories. --NeilN talk to me 17:07, 29 January 2015 (UTC)
Wikipedia is not a battleground of OR publishing academics (or any others). We summarize reliable secondary sources (and sometimes tertiary). Primary sources are not completely excluded either, but they are ought to be used with care. For example, a primary source on religion could be used if there are secondary sources supporting the very claim. Cheers! Jayaguru-Shishya (talk) 20:01, 30 January 2015 (UTC)
The issue, as noted above, is precisely what is reliable original research. Maybe the original poster hasn't seen, in ten years of editing, just how bad most of the original research that we see in Wikipedia is. Wikipedia doesn't have an infrastructure capable of peer review of original research. Wikipedia is an encyclopedia, not a peer-reviewed journal. We do get a lot of poor-quality original research, and we are able to reject it as original research; we don't have to spend months waiting for peer review to reject it as bad original research. A real academic scientist who has done high-quality original research knows not to submit it to Wikipedia but to a journal, and some journals do fast peer review for fast-breaking research. How would we, Wikipedia, determine what is reliable original research? It is better to have a policy against it than to have a lot of individual heated arguments about what is reliable original research. Robert McClenon (talk) 22:34, 29 January 2015 (UTC)
And even so. Peer review is to some extent the best system we have, but still lots of substandard (and even rubbish) research gets published often in low ranking but sometimes even in good journals. I have encountered these when making reviews or just reading up on a topic (and I am a regular reviewer for a host of highly rated scientific journals and am on the editorial board of one next to having published over 40 papers in scientific journals). So even the golden standard (as mentioned above) let's stuff slip through it shouldn't. However, in practice the damage is often limited as poor articles tend to be ignored; as on almost any topic many articles are written. (This is btw a good argument why scientific papers should often be considered a primary source but that is another discussion).
On Wikipedia, however, we aim to write only a single authorative article on each topic. If even peer review conducted by fellow scientists, personally known to editors cannot always guarantee high quality for original research; how can we aim for better here? Not a good idea to even try. Arnoutf (talk) 13:35, 30 January 2015 (UTC)
The OP is right - NOR leaves us impoverished. We can't look up census records, birth certificates, or marriage certificates to improve biographies. We can't quote from someone's unpublished letters that we find in an archive. We can't email a scholar to ask them "what did you really mean when you said x, x1 or x2?" We could do better work if we relaxed our NOR policy. The problem is, we would do it at a price. That price would be someone reading a couple sources and drawing a conclusion that's just plain wrong. That price would be endless debates about what a source really meant (or rather, even more endless debates, since they're easy enough to find, even now) with no good mechanism to shut them down. That price would be more people posting obviously crazy ideas. Guettarda (talk) 21:46, 30 January 2015 (UTC)
Vital records ARE NOT original research. They are primary sources. As our policies on primary and secondary sources explain, secondary sources are preferred, but reliable primary sources, and census records, birth certificates, and marriage certificates are reliable sources, do have their place. I strongly disagree with a statement that NOR leaves us impoverished. It does leave us without the "truth" resulting from reliable original research, but it also excludes a lot of really crazy ideas. That isn't impoverishment. At least, I don't think it is impoverishment. And we can cite primary sources when there are no secondary sources or when secondary sources disagree. Robert McClenon (talk) 22:14, 30 January 2015 (UTC)
One of my wife's great-great-great grandfather's marriage certificate is marked with an X with the notation "mark of [his name]". From that I can deduce that he was probably illiterate. Had their child been notable, that deduction might be interesting. Another one of her great-great-great-grandfathers (who is notable) was, according to his grandson's self-published memoir, married in Canada, but his wife was American. No published source will tell me what she was doing in Canada, but census records tell that her younger half-siblings were born in Canada, and that their father was Canadian. Mystery solved...not that it's interesting enough for this bio, but it's the kind of thing that can make sense of a life that sources don't bother to talk about. Finally, I recently found a photocopy of a letter from Susan B. Anthony to the same guy's father. He might be marginally notable, and if he were, this would complement nicely that published mentions I have found that mention him as a terminal 'stationmaster' in the underground railroad. Adding "women's rights supporter" to "anti-slavery activist" rounds a bio out nicely. But I can't use the letter.

This isn't a complaint. The benefits are 'well worth the cost. Guettarda (talk) 23:11, 30 January 2015 (UTC)

Robert McClenon, the OP has not been editing for ten years. I would state more, but the OP would likely be offended. Anyone is free to see why on my talk page. Flyer22 (talk) 21:54, 30 January 2015 (UTC)
Guettarda, we can email a scholar to ask them "what did you really mean when you said x, x1 or x2?." Editors have done that on Wikipedia, but we obviously can't use the scholar's words via email as a reference in the article. Flyer22 (talk) 22:03, 30 January 2015 (UTC)
This (KateWishing's "22:31, 29 December 2014 (UTC)" post) is a recent example of an editor having emailed a scholar for clarification. Flyer22 (talk) 22:10, 30 January 2015 (UTC)
Editors have done it. But if you say "I emailed [person] and they said it's actually x2", I'm going to say "not a reliable source" - mostly because I don't agree with you, of course. (If I already believed it was x2 we wouldn't be arguing about it.) Guettarda (talk) 23:12, 30 January 2015 (UTC)

I agree that it's become bad policy as far as it is typically interpreted these days and should go. And I'm basing that on my editing experience since 2005. It's become a short cut for people who feel that a conclusion or created impression is unfounded but can't come with an argument why. In the policy we read "...original research because it expresses a Wikipedia editor's opinion..." Yet everything Wikipedia says is said because editors are of the opinion Wikipedia should say that. It's in fact quite rare to encounter situations where WP:RS, WP:NPOV, etc would not be sufficient to address the problem AND you couldn't get a local consensus against the edit but for this policy. I am quite confident that the examples presented on this policy page could be quite readily dealt with simply by editors explaining why the implied conclusion is doubtful and getting consensus for that. As it stands now, the validity of the conclusion is irrelevant, just that it's made, as if Wikipedia isn't implying conclusions all the time. As I noted earlier on this page, "pretty much everything you do to an article... is going to "imply" some change to some "conclusion" somewhere to at least some degree, even if just at the overall article impression level." If the brakes were put on so that WP:OR were not used by deletionists with respect to editorial decisions to include material (so they'd just have "not RS" or "UNDUE" arguments available as trump cards), the policy would at least be circumscribed and accordingly might be salvageable. I have presented controversies and tried to dial down the hotly-expressed-dueling-opinion angle of "he said, she said" in favour of hewing closer to a less scandal-implying presentation of facts that may or may not imply controversy and been accused of OR. I then concede and present the battle just like it appears in the general media and I'm then told that we're not a general source, we're an encyclopedia, and therefore need to be more detached in tone. Well then, it's either one or the other. Either use sources just like the typical reliable source does, even if that typical reliable source happens to be low brow/mass market/attention-seeking, or we go the "more encyclopedic" way. If the latter, then the OR argument shouldn't be trotted out in order to argue for deletion of the Wikipedia-toned material. WP:OR at bottom reflects a lack of confidence that the Wiki concept works. An editor dares to think about the matter, and we can't handle that? In fact, we handle it all the time when people are reasonable enough to point out the problems with another editor's thinking instead of saying that the way the RS-cited material was put together has an element of originality as if precluding a substantive discussion of the dispute thereby solves it. As I note in the next section, a SYNTH charge is the most typical form of OR accusation as it has the best trumping power of this sort (calling a source primary is not as decisive).--Brian Dell (talk) 00:45, 8 February 2015 (UTC)

Definition of Research[edit]

Perhaps I'm missing something, but isn't all research by definition - unless plagiarized - original research? Further, if one relies on publishment as a means to legitimize research, what legitimizes a publisher? Surely not consensus - for myriad logical as well as ethical reasons. Just looking for clarification -- thanks. Amizzo (talk) 01:20, 7 February 2015 (UTC)

Original research is a term usually reserved for collecting, analyzing and interpreting primary data to gain answers to a thus far unanswered research questions. Almost all scientific research aims to be like that. Wikipedia is however not into primary scientific research but in aggregating existing knowledge.
A publisher can publish what they want using whatever quality standards they use. No consensus needed, and that is one of the reasons why scientific research is considered a primary sources in Wikipedia. Primary sources in themselves are fine, however without knowledge of the context it is hard to judge their contents at the right value.
However, the value of a journal depends highly on consensus in the academic field. Some journals are considered essential to a field, and those will be present in almost all scientific libraries, these are the journals scientists send their best work to, and these are the journals where publications contribute to the esteem of the author. Obviously such a journal gathers more revenue for a publisher.
If a respected journal lowers its standards, its desirability will drop over the years, number of good papers submitted to it will drop, etc. On the other hand new journals often need a number of years of uphill struggle to be accepted as a good journal. So there is definitely some consensus going on in determining what good journals are. (Of course an editor would need intimate knowledge of a field to know this). Arnoutf (talk) 09:03, 7 February 2015 (UTC)
On Wikipedia I think the term is more frequently used with respect to the collecting, analyzing and interpreting of secondary sources. The SYNTH section here doesn't mention "primary" even once, and most accusations of original research are accusations that someone has engaged in SYNTH. There is no absolute ban on using primary sources. There is on SYNTH. It follows that making out a SYNTH charge is more likely to get a decisive settlement in one's favour than accusing someone of using a primary source. Some people argue that there's a de facto ban on primary sources by making a big deal out of "secondary or tertiary sources are needed to establish the topic's notability" since when there are such secondary sources, there is no longer a need to use the primary source and the source could just be switched to the secondary. This notion of a double-cite RS policy (one cite for the material, and one cite for its "notability" even though the second cite would usually make the first cite redundant) is, however, not clearly and unambiguously WIkipedia policy (WP:RS doesn't indicate a need for two cites) unlike SYNTH hence a charge of SYNTH is the most common form of OR accusation.
That doesn't mean editors alleging OR won't try to run with whatever works, e.g. I add material to an article about an advocate of fringe theories like WiFi causes cancer, material I cite to secondaries that challenge the fringe theory, and I get accused of SYNTH because I've caused readers to think the advocate might have an axe to grind and the cited sources don't criticize the advocate directly, just the theory. I then try to accommodate by substituting in direct criticism of the advocate cited to a widely cited critic but use the critic as a primary. I'm then called upon to produce a secondary source in order to show that the controversy is "notable" (both because WP:OR says "secondary or tertiary sources are needed to establish the topic's notability" and the now more direct criticism means there's supposedly a BLP issue that in turn requires "multiple reliable third-party sources documenting the allegation or incident"). Well the reason why secondary sources covering the "controversy" are hard to come by is because the advocate is too fringe to get equal time with the critics or, more to the point, get the attention of critics! You end up with a Wikipedia with all sorts of article subjects who are notable enough to get an article but not notable enough to get secondary source coverage of their deviation from the mainstream that explicitly expresses it as such. People read these stub articles and reckon the subject is every bit the authority the subject claims to be, use of sources other than the subject being so sharply circumscribed. The way WP:OR is often used thus frequently enables the very quackery it's supposed to limit. WP:OR nonetheless makes perfect sense if one believes most of the dubious information in Wikipedia did not originate off-Wiki or with editors not following WP:RS and WP:NPOV but right here on-Wiki with WP:RS and WP:NPOV-following editors who synthesize secondary sources or use primary sources.--Brian Dell (talk) 03:13, 8 February 2015 (UTC)

Is WP:Cherrypicking#Contradictions invalidated by WP:STICKTOSOURCE?[edit]

WP:Cherrypicking#Contradictions says, "As to contradictory information, if, for example, a source says 'Charlie loves all blue coats and hates all red coats', to report in Wikipedia that according to the same source "Charlie loves all ... coats" is cherrypicking. It is cherrypicking even if the source is precisely cited."

WP:STICKTOSOURCE says, "Take care not to go beyond what is expressed in the sources, or to use them in ways inconsistent with the intention of the source"

Suppose a source itself baldly commits exactly the example error by saying "Charlie loves all blue coats and hates all red coats; therefore Charlie loves all coats." To avoid cherrypicking, it would be necessary to report on the contradiction. However, the intention of the source is to prove that Charlie loves all coats, so it would not be sticking to the source to mention that Charlie hates red coats.

Sticking to a source belongs to WP:OR, so would seem to be more vital than the essay on Cherrypicking. More vital still, though is WP:V. Something that is verifiably false from a source is not verifiable from that same source. WP:EP also says, "a lack of information is better than misleading or false information". A source like this is unusable if the contradiction can't be raised.

"Charlie hates all red coats" might be an important claim for an article, though. It would be a different instance of cherrypicking to use it without mentioning the source's spurious conclusion of loving all coats, but I think as long as that's mentioned, the source's intent should not prevent the citation of Charlie hating red coats. STICKTOSOURCE just needs to be interpreted less encompassingly. While the source intended to conclude that Charlie loves all coats, it also intended to say Charlie hates red coats at the point that it said it.

WP:NOTOR has some things to say about handling two contradictory sources that I think apply equally well to a single source contradicting itself: "We have a responsibility to present an accurate and factual overview of the topic addressed in the article. This may include indicating when a given authority may be wrong and presenting contradicting claims using proper weighting. A solution is to accurately and honestly cite the authority, while also citing the conflicting fact(s); point to the problem, but do not attempt to solve it with your own arguments." Rhoark (talk) 02:48, 14 March 2015 (UTC)

Suppose a source itself baldly commits exactly the example error by saying "Charlie loves all blue coats and hates all red coats; therefore Charlie loves all coats." To avoid cherrypicking, it would be necessary to report on the contradiction. No, it's not necessary. A contradiction conveys no useful information; any part of a source that contradicts itself is clearly unreliable. So simply don't use it at all. This isn't "cherrypicking" which is selective use of sources intended to mislead. Peter coxhead (talk) 09:10, 14 March 2015 (UTC)
To clarify, its not a priori necessary to report on the contradiction, but if any of the three prongs (loves blue, hates red, loves all) is reported, it would be cherrypicking unless all three are reported. Rhoark (talk) 13:45, 14 March 2015 (UTC)
Precisely. Peter coxhead (talk) 17:34, 14 March 2015 (UTC)