Wikipedia talk:No original research
![]() | This page is only for discussions about the Wikipedia page Wikipedia:No original research. To discuss an article, please use that article's talk page. To ask for help with using and editing Wikipedia, use our Teahouse. Alternatively, see our FAQ. | This page is not meant for general questions, nor discussions about specific articles.
![]() | 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, and don't panic. |
If you want to know whether particular material constitutes original research or original synthesis, please use the No original research notice board. Questions about the policy itself may be posted here. |
|
Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 51, 52, 53, 54, 55, 56, 57, 58, 59, 60 61, 62, 63, 64 |
|
This page has archives. Sections older than 90 days may be automatically archived by Lowercase sigmabot III when more than 3 sections are present. |
When will Wikipedia start allowing original research?
When will Wikipedia start allowing original research? 2601:646:8B00:D590:A8D8:A7CD:CD01:F7C5 (talk) 06:38, 12 December 2023 (UTC)
- It's unlikely that we ever will.—S Marshall T/C 08:47, 12 December 2023 (UTC)
- There's no good reason for us to do that, and it's not consistent with being an encyclopedia. If you have something new that you want to tell the world, post on social media, or start your own website, or send it to an academic journal, or write a book. Ideas and information that the world has never seen before belong in those places, not here. WhatamIdoing (talk) 15:39, 12 December 2023 (UTC)
- In great line I'm agree.
- But I can imagine some ones can be still confused.
- For example:
- Someone is an emeritus professor.
- He or she is a specialist in what he or she knows.
- If I understand correct you, he or she may put his or her facts on Wikipedia what he or she knows and that is published. But he or she is not allowed some facts he or she knows wich isn't published yet. Shy Aroace (talk) 08:30, 24 January 2024 (UTC)
- Correct. Wikipedia is not the right venue for presenting facts that have not yet been published. A professor emeritus (a specialist in her field) would know this - and would also know what the correct venues to publish new information are. Blueboar (talk) 12:35, 24 January 2024 (UTC)
- It’s unlikely that someone in that position would wish to share unpublished research, and it would likely go against the policy of any university or paper that that research is meant to be published by.
- Someone in that position would be aware that if they’re the first person making the claim that they will be expected to provide evidence.
- Them simply saying that the evidence is in unpublished research is an issue because it both asserts something without a reliable source or verifiability, and that it asks other editors to trust their claim based on their position.
- If the evidence will be published then adding it to Wikipedia can wait. If it isn’t publishable for some reason then it shouldn’t be here in the first place. Cbrfield (talk) 04:12, 2 April 2024 (UTC)
- There's no good reason for us to do that, and it's not consistent with being an encyclopedia. If you have something new that you want to tell the world, post on social media, or start your own website, or send it to an academic journal, or write a book. Ideas and information that the world has never seen before belong in those places, not here. WhatamIdoing (talk) 15:39, 12 December 2023 (UTC)
- Actually I changed my mind on this. I think we will start allowing original research within the next 5 years—with some possible estimates being as short as 100 hours until the ban on original research is finally lifted. Remsense诉 04:15, 2 April 2024 (UTC)
Using OR to omit information
The policy is clear that we can not use our own original research to state something in an article. What is less clear is whether we can use our own original research to omit something from an article. As an example: say an article says X occurred, cited to several (reliable) newspapers reports saying that X occurred. But an editor digs a bit, and discovers that X didn't actually occur… that the papers got the story wrong. That editor can’t add “However, X didn’t occur, and the papers got it wrong” based on his original research, but he CAN discuss his research on the article talk page and gain a consensus to omit the statement that X occurred. Should we mention this in the policy… or is the situation rare enough that spelling it out in the policy would just confuse people? Blueboar (talk) 13:38, 24 February 2024 (UTC)
- We probably should put the big unspoken reality / need into words. Which is "Wikipedia editors are editors" and then that would be one of the zillions of types of situations covered by it. :-) North8000 (talk) 14:46, 24 February 2024 (UTC)
- Wouldn't that be a reliability issue? The best sources are still only generally reliable, they can be wrong. -- LCU ActivelyDisinterested «@» °∆t° 15:25, 24 February 2024 (UTC)
- I do think that editors should be encouraged to use their own WP:Editorial discretion. (I also think we need another policy statement, probably in WP:BALANCE, that says editors should be writing an encyclopedia article.)
- The problem that I foresee is illustrated at Wikipedia:Reliable sources/Noticeboard#MEDRS sources on curcumin supplementation. If you look towards the end of that section, I've evaluated one of the disputed sources against the actual criteria. It turned out to be an ideal: systematic review in a top-rated journal, no COIs, etc. Given the person who reverted it, I can make a good guess about what the real problem is: it said something somewhat positive about a substance that isn't already approved by the FDA as a regulated drug. Many of our anti-woo warriors (for whom I truly am grateful; Wikipedia would have worse content and more spam without them) would cheerfully use their editorial discretion to blank anything that gives the "wrong" answer about an unregulated dietary supplement. I assume that POV pushers of all sorts would do the same: I think all the sources saying that X happened are wrong, so we should omit them all: no Moon landing, Elvis didn't die, etc.
- On the one hand, I don't believe that any WP:VOLUNTEER should be forced to add a source that they suspect is incorrect, but I also don't want to say that they can remove it (and force others to accept that removal). If you have a consensus to omit, then consensus should prevail; if you have a consensus to include (despite your own objections), then you should let one of the supporters add it to the article.
- If you need a simple example for discussion, the OED regularly provides information about the first known use of a word in writing, and a trip to books.google fairly often proves it wrong (for entries that haven't been updated for a while). The uncontroversial solution in such cases is to silently omit the OED's claim. WhatamIdoing (talk) 03:26, 28 February 2024 (UTC)
- I agree that we need to allow editors discretion. Verifiability is necessary but not sufficient for inclusion, and there are valid reasons to exclude information that can be verified. On rare occasion, there may be multiple different reports in multiple different sources, and it may make more sense to omit the contradiction than to report it. Or it might simply be too tangential or detailed for a summary style article. Deciding that something isn't relevant is the essential job of an editor. Shooterwalker (talk) 14:31, 29 February 2024 (UTC)
- Yes, and there are degrees of WP:Relevance which enter into those decisions. North8000 (talk) 15:52, 29 February 2024 (UTC)
- I agree that we need to allow editors discretion. Verifiability is necessary but not sufficient for inclusion, and there are valid reasons to exclude information that can be verified. On rare occasion, there may be multiple different reports in multiple different sources, and it may make more sense to omit the contradiction than to report it. Or it might simply be too tangential or detailed for a summary style article. Deciding that something isn't relevant is the essential job of an editor. Shooterwalker (talk) 14:31, 29 February 2024 (UTC)
Reverted edit of 3 respected sources that confirm the subject of the article was a terrorist.
Is there a consensus to revert to the edit I made this evening? These are the sources.
https://www.bbc.com/news/extra/xkbwldvmb5/exposed-the-secret-army https://m.belfasttelegraph.co.uk/video-news/video-footage-of-martin-mcguinness-showing-kids-bomb-and-gun/38473261.html https://www.newsletter.co.uk/news/politics/new-footage-shows-martin-mcguinness-building-car-bomb-and-giving-bullets-to-kids-943764 Jaymailsays (talk) 20:21, 27 March 2024 (UTC)
- Complete misrepresentation of the objection. You changed several existing sentences;
- In response, McGuinness said the statements were "fantasy", while Gearóid Ó hEára (formerly Gerry O'Hara), a Derry Sinn Féin councillor, stated that he and not Ward was the Fianna leader at the time
- This was changed to;
- The terrorist act, injured at least 26 people and cast real doubt on Gearóid Ó hEára (formerly Gerry O'Hara), a Derry Sinn Féin councillor, statement that it was he and not Ward who was the Fianna leader at the time, the "incendary footage" of McGuinness proves that the Inquiry was misled
- The second sentence originially read;
- The inquiry concluded that, although McGuinness was "engaged in paramilitary activity" at the time of Bloody Sunday and had probably been armed with a Thompson submachine gun, there was insufficient evidence to make any finding other than they were "sure that he did not engage in any activity that provided any of the soldiers with any justification for opening fire"
- You changed to this to;
- The inquiry, based on the evidence available at the time, concluded that although McGuinness was "engaged in paramilitary activity" at the time of Bloody Sunday and had probably been armed with a Thompson submachine gun, there was insufficient evidence to make any finding other than they were "sure that he did not engage in any activity that provided any of the soldiers with any justification for opening fire"
- As I already said, there isn't any dispute McGuinness was in the IRA at the time. But you can't use the existence of a film made several weeks after Bloody Sunday (1972) to try and cast doubt on the findings of the Bloody Sunday Inquiry. Kathleen's bike (talk) 20:38, 27 March 2024 (UTC)
- This page is for discussing changes to the NPOV policy, I would suggest taking this to the talk page of the article in question. But one point to remember is that just because something can be verified doesn't guarantee inclusion, rather inclusion requires that something can be verified. -- LCU ActivelyDisinterested «@» °∆t° 20:55, 27 March 2024 (UTC)
This is the wrong place for this discussion. This talk page is for discussion of improvements to the WP:No original research page and policy. The best place to start your discussion is at the talk page of the article that you are writing about. There are ways and places to take it farther after that if needed. North8000 (talk) 20:54, 27 March 2024 (UTC)
- Alternatively, @Jaymailsays might find it more appropriate to take the question to Wikipedia:No original research/Noticeboard or Wikipedia:Neutral point of view/Noticeboard. It sounds to me like this is more of an NPOVN question (i.e., some sources cast doubt on the inquiry's findings, and other sources do not), but if @Kathleen's bike really believes the main problem is adding up multiple sources to create material that isn't in any source at all ("SYNTHESIS"), then NORN would be the better choice. WhatamIdoing (talk) 23:27, 27 March 2024 (UTC)
- No, it's not a NPOV issue. None of the three sources above cast any doubt on the Bloody Sunday Inquiry findings, the text added saying the "incendary footage" of McGuinness proves that the Inquiry was misled is Jaymailsays's own constructed narrative, no sources say that at all. As I posted at Talk:Martin McGuinness#Synthesis and false narrative Martin McGuinness's own evidence to the inquiry in 2003 has him involved in the IRA at that time and covers the bombing campaign directed at businesses in Derry, so this video (known about since 2019) is, if anything, consistent with his testimony not casting any doubt on it. Kathleen's bike (talk) 00:27, 28 March 2024 (UTC)
Feedback requested on whether usage examples require sources
Your feedback would be welcome at this discussion about usage examples at linguistic articles, and whether they require sources: Talk:Franglais#Are usage examples "original research"? Thanks, Mathglot (talk) 17:14, 22 April 2024 (UTC)
for WP:SYNTH, is semantic change relevant?
if the socially recognised term for an article has been coined after a source had been posted (or before the term was popularised), is that source valid or is that WP:SYNTH? for example, if an article is detailing lemons, but does not mention 'lemon' by name because the word (hypothetically) had not been invented yet or was extremely novel and esoteric at the time, is that source a valid source? NotQualified (talk) 21:06, 7 May 2024 (UTC)
- i would argue there is precedent WP:TRANSCRIPTION, as sources from different languages refer to the same idea but with different terminology yet still can be sourced. we need to outline clearly in the rules whether semantic change is or isnt covered by WP:SYNTH. basically, does it invalidate a source if the term was not used in verbatim? NotQualified (talk) 21:09, 7 May 2024 (UTC)
- also, how far can semantic change be tolerated? what if the source being relevant to a topic could be highly likely but still with reasonable doubt? are only sources that are irrefutably discussing a topic without mentioning it by name acceptable, if even any are? or are sources beyond reasonable doubt also okay? NotQualified (talk) 21:14, 7 May 2024 (UTC)
- My take on this, in context, is this:
- The consensus among Wikipedia editors is that certain terms, such as the Great Replacement and Cultural Marxism, refer in their plain meaning to what the best sources available understand to be conspiracy theories. The articles at those titles are therefore articles about those conspiracy theories.
- Where there are real-life phenomena or concepts that the conspiracy theorists appropriate for their political fictions, like White demographic decline and Marxist cultural analysis or the Long march through the institutions, enwiki has separate articles on those topics. What we avoid doing is incorporating real-world phenomena into articles about conspiracy theories in a way that might lend credence to the fictions (and in this we follow WP:FRINGE). Newimpartial (talk) 00:21, 8 May 2024 (UTC)
- whats your take out of context, in a more general sense. it still needs to be defined in rules if semantic change is acceptable NotQualified (talk) 10:06, 8 May 2024 (UTC)
- No it doesn't. There's no such thing as a rule about language "out of context". Remsense诉 10:28, 8 May 2024 (UTC)
- no i mean, if we had to clarify if semantics is or isnt convered. ignoring the context of where i initially was stumped by the allegation it wasnt and i broke rules and rather just thinking in terms of wikipedia. does WP:SYNTH disqualify semantic change affected sources. if not, this needs to be clarified NotQualified (talk) 10:43, 8 May 2024 (UTC)
- So, we're writing an encyclopedia, which is a tertiary source. What that means here is, if the meaning of language has significantly shifted since a source has published, it's often not a source we rely upon when writing an encyclopedia. For example, we consider many ancient sources like the Book of Han to be primary sources to be used with caution, even if we certainly can't omit them entirely from some articles. We rely on more recent secondary sources to put those primary sources in context. So in your initial example, if no other source includes the earlier mention of lemons, it's probably not within the scope of what we're writing about. We reflect the body of sources about a topic, which usually does a good job of clearly defining the thing we're writing about. Remsense诉 10:53, 8 May 2024 (UTC)
- no i mean, if we had to clarify if semantics is or isnt convered. ignoring the context of where i initially was stumped by the allegation it wasnt and i broke rules and rather just thinking in terms of wikipedia. does WP:SYNTH disqualify semantic change affected sources. if not, this needs to be clarified NotQualified (talk) 10:43, 8 May 2024 (UTC)
- To answer your question: I don't think we have a rule (about semantic drift, say) that applies universally on enwiki regardless of context. What we do have - starting with WP:N and WP:PAGEDECIDE - is a set of rules, guidelines and practices about topics and the scope of articles. Where reliable sources treat terms as synonyms - like the Frankfurt School conspiracy theory as a synonym for the Cultural Marxism conspiracy theory - our articles treat them as the same topic. When reliable sources distinguish between the meanings of terms - like Social justice and Political correctness - our articles distinguish them. And our articles are, broadly speaking, supposed to follow reliable sourcss in defining the boundaries of a topic.
- As an aside, I have frequently seen "it doesn't mention the article title!" used as a lazy reason to exclude a source from an article, which is not grounded in policy. In many cases, however, this argumentation is shorthand for another argument that is based in policy, namely, that the statement for which the source would be used is out of scope for the article for which it is proposed and it would be a SYNTH violation to include it. Whether a source is usable or not, however, doesn't depend on the presence or absence of a keyword but rather on whether the statements for which it is reliable fall within the scope of the article for which it is proposed. Newimpartial (talk) 11:11, 8 May 2024 (UTC)
- I think it's usually pretty clear when writers are talking about the same thing because they'll usually mention each other. Remsense诉 11:14, 8 May 2024 (UTC)
- im having a disagreement with a writer for if a source is related to the article and they dismissed it saying the title wasnt mentioned verbatim and id inevitably get banned for putting it in. they did correctly remove a bit i wrote that was unsourced, but that was due to me not adding the source in the article but in the talk page which on reflection was massively stupid lmao. NotQualified (talk) 12:45, 8 May 2024 (UTC)
- stupid of me
- NotQualified (talk) 12:46, 8 May 2024 (UTC)
- I'm not sure what precisely you're referring to, but a constantly recurring problem with articles like Cultural Marxism conspiracy theory and Great Replacement is that editors arrive and try to impose their BOTHSIDES interpretation of NPOV, where one of the "sides" has no significant support among WP:RS. What typicallly happens is that either those editors will seek the inclusion (for "facts") of what are at best WP:RSOPINION sources, or they will do their own "research" into a topic the conspiracy theory is interpreting and seek inclusion of related statements in the article, as if to say, "here is the actual phenomenon that the conspiracy theory is interpreting".
- The problem with the latter is the inevitable WP:SYNTH violation - unless reliable sources actually tie the conspiracy theory back to real world phenomena (usually by contrasting one with the other), editors are forbidden from doing so in articles. What is more, even if one source does this, it may be WP:UNDUE to mention in article space if the vast majority of sources do not see a clear connection between thr conspiracy theory's claims and actual phenomena. And WP:FRINGE tells us to be very careful about allowing wikipedia articles to become amplifiers for fringe claims, including those by conspiracy theorists. So while the key question is not actually, "does the source name the article topic?", there are lots of sourced statements that would be completely inappropriate to include in articles - especially articles on FRINGE topics - because of SYNTH concerns. Newimpartial (talk) 13:35, 8 May 2024 (UTC)
I'm not sure what precisely you're referring to
NQ is referring to this discussion: [1]. --Hob Gadling (talk) 15:47, 12 May 2024 (UTC)
- how did you get the cool text around your usrrname? NotQualified (talk) 12:47, 8 May 2024 (UTC)
Check out WP:CUSTOMSIG. I think mine is exactly the maximum allowed size—because I had to use a lot of characters for a
<span>...</span>
tag to properly tag the Chinese character as being in Chinese. Remsense诉 12:52, 8 May 2024 (UTC)
- im having a disagreement with a writer for if a source is related to the article and they dismissed it saying the title wasnt mentioned verbatim and id inevitably get banned for putting it in. they did correctly remove a bit i wrote that was unsourced, but that was due to me not adding the source in the article but in the talk page which on reflection was massively stupid lmao. NotQualified (talk) 12:45, 8 May 2024 (UTC)
- I think it's usually pretty clear when writers are talking about the same thing because they'll usually mention each other. Remsense诉 11:14, 8 May 2024 (UTC)
- No it doesn't. There's no such thing as a rule about language "out of context". Remsense诉 10:28, 8 May 2024 (UTC)
- whats your take out of context, in a more general sense. it still needs to be defined in rules if semantic change is acceptable NotQualified (talk) 10:06, 8 May 2024 (UTC)
- also, how far can semantic change be tolerated? what if the source being relevant to a topic could be highly likely but still with reasonable doubt? are only sources that are irrefutably discussing a topic without mentioning it by name acceptable, if even any are? or are sources beyond reasonable doubt also okay? NotQualified (talk) 21:14, 7 May 2024 (UTC)
When a term is coined isn't usually relevant. Actually the semantics rarely matter a lot, on Wikipedia -- it's really the meaning that matters. Our articles have to mean what the sources mean, but they don't have to use the sources' wording. However, if the purpose of your post here is to develop an argument about Great Replacement or other topics related to the far right or alt right, then my advice would be to leave those topics alone for the time being. Editing them isn't usually a pleasant or productive thing to do, even for very experienced editors, and for newer people they're no fun at all.—S Marshall T/C 22:07, 7 May 2024 (UTC)
- i am very new to this and youre right it's exhausting. it's fierce and obvious ideologically blockading. im a leftist but im able to be objective to anything yet im told that i cant edit the theory in any way that 'launders' it, even when im sourcing objective facts and quotation. they are biased. however, no i am not just here for that. i genuinely do think semantics need to be addressed within this rule as this confusion has to be quashed on if it's sound or not. if group consensus says it's fine, im going back to edit it. if not, i'll find new sources. NotQualified (talk) 00:05, 8 May 2024 (UTC)
- thank you for your welcomingness marshall NotQualified (talk) 00:06, 8 May 2024 (UTC)
- off-topic, if a reliable source sources an unreliable source, do i take it that the reliable source fact-checked the unreliable one? NotQualified (talk) 00:09, 8 May 2024 (UTC)
- No. Sources aren't just "reliable" or "unreliable"; they're a continuum, where some sources are more reliable than others. Generally speaking, the pinnacle of reliability is an academic meta-analysis or systematic review. Less reliable, but still very reliable, is a peer-reviewed study in a scholarly journal. News articles aren't brilliantly reliable but might be the best available for some topics, in which case I'd prefer a recent or current news article published by a newspaper of record. Where a seemingly-reliable source turns out to depend on an unreliable source, then we'd usually say that's a consideration that makes the seemingly-reliable source more unreliable. But be careful. Make sure that, for example, the reliable source isn't quoting the unreliable one in order to debunk it.—S Marshall T/C 08:30, 8 May 2024 (UTC)
- right, thanks. lots of info to take in. i'll make a talk page discussion if the semantics argument is approved in rules and have you review my writing to ensure no rules are broken, if you wouldnt mind NotQualified (talk) 10:04, 8 May 2024 (UTC)
- i feel like we have broad consensus now on what policy should be, would you like to edit this article to mention this and i close this topic? NotQualified (talk) 12:49, 8 May 2024 (UTC)
- I feel this isn't too necessary an addition, as we're mostly just summarizing how we already understand the policy to be. Generally, less is more with policy—you want the minimum possible verbiage with enough flexibility but also so that no one can ever get anything twisted in an argument. But it's always worth asking what other editors think! Remsense诉 12:54, 8 May 2024 (UTC)
- No. Sources aren't just "reliable" or "unreliable"; they're a continuum, where some sources are more reliable than others. Generally speaking, the pinnacle of reliability is an academic meta-analysis or systematic review. Less reliable, but still very reliable, is a peer-reviewed study in a scholarly journal. News articles aren't brilliantly reliable but might be the best available for some topics, in which case I'd prefer a recent or current news article published by a newspaper of record. Where a seemingly-reliable source turns out to depend on an unreliable source, then we'd usually say that's a consideration that makes the seemingly-reliable source more unreliable. But be careful. Make sure that, for example, the reliable source isn't quoting the unreliable one in order to debunk it.—S Marshall T/C 08:30, 8 May 2024 (UTC)
- Editors need to Wikipedia:Use our own words. If a word is unclear (e.g., because people from different generations or English varieties understand it differently), then use a clearer word, or add an explanation. I find that editors are more likely fight over the article titles ("No, my diet is the One True™ Ketogenic diet, not this rare thing for kids with epilepsy that's been called by that name for more than a century!") but not individual words. When you do get someone who fights over clear writing, you may be dealing with an editor who seems to believe that if it's not a copyvio (or at least plagiarism), then it's a NOR violation. Resolving that usually takes some work and sometimes some help, but it's usually achievable. WhatamIdoing (talk) 00:05, 11 May 2024 (UTC)
Election Percentages
There was a Presidential Election in North Macedonia on 8 May and I and StephenMacky1 are in dispute regarding the manner in which the percentages of votes should be displayed in the Lead Infobox. This is because the source is The Electoral Commision, which reports percentages by dividing by the total number of all votes (including blanks). Before we get into a giant dispute, I would wish to have input from those more familiar with working on such articles.
I maintain that we should use percentages of valid votes in the Lead Infobox, which is how identical issues were handled in the previous election (compared with page 32 here). I inserted hidden text into the infobox with that reasoning and restored the percentages to the fraction of the total without blanks.
StephenMacky1 absolutely disagrees and reverted things for the current version of the page. They maintain that we should report it EXACTLY as the sources do and that the other election does not constitute a precedent or even implied consensus, but is in fact an indicator that the article is itself in error by not abiding due to not abiding by the sources perfectly. This also causes an internal disagreement on the page itself, because the results use Template: Election results that do not allow us to use blank votes as a valid part of their calculation and mean that the table refuses to display anything, but the percentage of effective votes contrary to StephenMacky1's wishes for percise reporting of what the sources state.
Does anybody know if the Manuel of Style addresses this? I have not been able to find such a specific rule, but I may have not been thorough enough. Is No original research so broad that, recalculating the numbers in the way the Template: Election results already does and placing that in the infobox at the top, would constitute a violation?
Also how does one call StephenMacky1 into this conversation, since I have previously not done this? Do I just request their presence on their talk page? 212.79.110.147 (talk) 21:35, 10 May 2024 (UTC)
- Recalculating numbers doesn't sound like a good idea. Gawaon (talk) 22:16, 10 May 2024 (UTC)
- Calculation is not a matter of style. This is a WP:NOR issue. NebY (talk) 22:33, 10 May 2024 (UTC)
- It is now also placed for discussion at their talk page. See: Wikipedia talk:No original research#Election_Percentages 212.79.110.147 (talk) 22:42, 10 May 2024 (UTC)
- @NebY: This is pretty clearly allowed under WP:CALC. Hey man im josh (talk) 22:54, 10 May 2024 (UTC)
- Whether it is or not, this is the wrong place. WP:CALC is a section of WP:NOR, not a style matter. NebY (talk) 23:07, 10 May 2024 (UTC)
- Calculating a percentage is not "research", original or otherwise. It is presenting information, entirely within scope for wikipedia Newystats (talk) 12:37, 24 May 2024 (UTC)
(This was placed here on suggestion at the Manual of Style talk page that it should be discussed here) 212.79.110.147 (talk) 22:41, 10 May 2024 (UTC)
- NOTICE: StephenMacky1 has been invited. User:Number 57 has edited the page for the election to use percentages of the total vote again at 22:56, 10 May 2024 (UTC). I have invited him here at 23:50, 10 May 2024 (UTC).
- I've put a pointer to this discussion at Wikipedia talk:WikiProject Elections and Referendums. NebY (talk) 23:34, 10 May 2024 (UTC)
- Thanks for posting that, NebY.
- 212.79, both of these approaches are okay under our policy. The ideal approach is likely to be naming both, i.e., "65% of all votes cast and 69% of valid votes". You could also add an Wikipedia:Explanatory footnote (using the {{efn}} template, if you'd prefer) to provide that information. WhatamIdoing (talk) 00:13, 11 May 2024 (UTC)
- Hello IP. Thank you for informing me. I was not aware of this change by the Electoral Commission. It was not necessary for this issue to appear on multiple talk pages. I agree with the suggestion above. A note could be added to explain how the percentages were calculated. StephenMacky1 (talk) 09:00, 11 May 2024 (UTC)
- There's no requirement to use {{Election results}} and we often don't. Why not present the results in the body of the article and the infobox in a way that accords with the source, e.g. by using the {{Election box}} family in the body? NebY (talk) 09:04, 11 May 2024 (UTC)
- Is there any kind of general preference between these possibilites for displaying it? I am personally biased toward Version 3, but will defer to senior editors if there are strong feelings against it.
|
|
|
|
212.79.110.147 (talk) 14:23, 11 May 2024 (UTC)
- Version 4, the percentages declared by the Electoral Commission, being of the "Total number of voters who voted" i.e. turnout. The body of the article should also present that percentage. However, I can't see what the basis is for calling the invalid votes "blank"; the English-language version of the page says "Total number of invalid ballots" and Google Translate gives the same. Why not call them "invalid"? I'd add that I am in sympathy with the Electoral Commission's choice, given that a formidable 5.6% of the ballots were invalid. NebY (talk) 16:05, 11 May 2024 (UTC)
- I called them blank, because I forgot North Macedonia doesn't distinguish Invalid and Blank votes like the 2024 Panamanian general election, which I have also edited recently, does. I have altered the above version accordingly.212.79.110.147 (talk) 18:22, 11 May 2024 (UTC)
- Thanks. It turns out that we can use {{Election results}} and still produce candidates' percentages as a % of turnover, per the North Macedonia Election Commission. I've adapted the article's table at User:NebY/sandbox/election. NebY (talk) 19:35, 11 May 2024 (UTC)
- I called them blank, because I forgot North Macedonia doesn't distinguish Invalid and Blank votes like the 2024 Panamanian general election, which I have also edited recently, does. I have altered the above version accordingly.212.79.110.147 (talk) 18:22, 11 May 2024 (UTC)
This is more of an infobox question than whether the calc is forbidden by WPNOR. In cases like this, explanation needs to be included. Sometimes that means leaving it out of the infobox and just covering in in the body of the article. North8000 (talk) 15:10, 11 May 2024 (UTC)
- Just over a week ago we had local elections in the UK. I have checked the official sources for some of the results, and I find that the percentages take invalid votes into account - so for example with this result, the only way that you can reconcile the 25.82% turnout is by adding the 13 rejected ballot papers to the (469 + 220 + 936 + 187 + 121) = 1933 valid votes: (1933 * 100 / 7537) = 25.65% but ((1933 + 13) * 100 / 7537) = 25.82%, as shown on the results page. This convention will not necessasarily be observed elsewhere. --Redrose64 🌹 (talk) 15:27, 11 May 2024 (UTC)
- This discussion concerns the percentage assigned to the candidate rather then turnout numbers. I have always seen turnout reported for all votes including invalid and blank one and am in agreement with you there. I don't think there even exist an election, where the turnout is reported in a different way.
- I also noticed that the Cherwell Council page over here would seem to be doubly incorrect according to your proposition that all votes be used as per the source. The percentages reported there, at the moment at least, divide by the valid votes (1933) and not all the votes (1946), since Matt Hodgson has 48.4% and not 48.1% next to him, and report the turnout as 1933 in absolute numbers, while having the percentage set to 25.82%.212.79.110.147 (talk) 18:22, 11 May 2024 (UTC)
- It was Icc27 (talk · contribs) who added the percentages there, I don't know what their source was. --Redrose64 🌹 (talk) 18:52, 11 May 2024 (UTC)
- It would be normal to display turnout figures including the rejected votes, but for candidate percentages to exclude these.
- As for reporting the turnout figure in terms of number of ballots I think it would be more normal to include those rejected, but I'm not aware of a general standard in this case. Icc27 (talk) 21:21, 11 May 2024 (UTC)
- Yes, you are correct. Standard election results presentation is for candidate/party percentages to be calculated using valid votes, and turnout to be based on all votes cast (valid and invalid). Some electoral commissions publish candidate/party percentages that have been calculated by including invalid votes (or in some cases excluding certain types of valid votes) but this is usually because their electoral law says that the calculation of the electoral threshold (for winning seats or whether a second round is required) includes invalid votes. Cheers, Number 57 03:52, 12 May 2024 (UTC)
- Agree. To add: I would say that it is fine but not necessary to include invalid votes in percentage calculations if (and only if) the number of invalid votes is specifically published, but undervotes should virtually never be included in percentage calculations. A topline election may have a 2% undervote while a simultaneous downballot election has a 20% undervote, but they should be treated the same. The undervotes would never (to my knowledge) be included in percentages for races with high undervotes, so they should not be included for races with low undervotes. Thus, invalid votes should only be included in percentages if they are distinct from undervotes. Star Garnet (talk) 06:49, 24 May 2024 (UTC)
- Yes, you are correct. Standard election results presentation is for candidate/party percentages to be calculated using valid votes, and turnout to be based on all votes cast (valid and invalid). Some electoral commissions publish candidate/party percentages that have been calculated by including invalid votes (or in some cases excluding certain types of valid votes) but this is usually because their electoral law says that the calculation of the electoral threshold (for winning seats or whether a second round is required) includes invalid votes. Cheers, Number 57 03:52, 12 May 2024 (UTC)
- It was Icc27 (talk · contribs) who added the percentages there, I don't know what their source was. --Redrose64 🌹 (talk) 18:52, 11 May 2024 (UTC)
It appears the same conflict is now present at the Lithuanian Presidential Election from 12 May, although it won't effect the first infobox just yet until the second round in two weeks. They also report their results of percentages of the total vote at their results page. There might need to be some sort of official ruling or this might cause an identical editing conflict there. 212.79.110.147 (talk) 09:11, 13 May 2024 (UTC), user accessed the page from a computer at 88.146.220.100
SYNTH: imply vs infer
WP:SYNTH says do not combine different parts of one source to reach or imply a conclusion ...
. Shouldn't that be "infer a conclusion"? RoySmith (talk) 19:29, 4 June 2024 (UTC)
- Since this is giving advice to writers, "imply" is correct. A writer implies conclusions and a reader infers them. Firefangledfeathers (talk / contribs) 19:32, 4 June 2024 (UTC)
- But, I was thinking of the editor reading the source and drawing inferences from different parts. RoySmith (talk) 20:37, 4 June 2024 (UTC)
- I'm fine with editors reading sources and inferring conclusions based on different parts as long as they don't write article text that implies those same conclusions. Firefangledfeathers (talk / contribs) 20:40, 4 June 2024 (UTC)
- I agree, but in that case, "reach" also needs to either be changed (to "state" for instance) or removed. M.Bitton (talk) 20:43, 4 June 2024 (UTC)
- Yes, "imply" is correct but "reach" is ambiguous here, as it can refer either to the editor reaching that conclusion for themselves from their reading of the source or reaching it for the reader in their own writing. Either remove "reach or" or, perhaps better, replace "reach" with "lead to". Largoplazo (talk) 20:50, 4 June 2024 (UTC)
- How would you feel about M.Bitton's suggestion of "state"? Firefangledfeathers (talk / contribs) 02:18, 5 June 2024 (UTC)
- I'm fine with editors reading sources and inferring conclusions based on different parts as long as they don't write article text that implies those same conclusions. Firefangledfeathers (talk / contribs) 20:40, 4 June 2024 (UTC)
- But, I was thinking of the editor reading the source and drawing inferences from different parts. RoySmith (talk) 20:37, 4 June 2024 (UTC)
I'm not Largoplazo but the reason I prefer "reach", or even "suggest" is:
- Source A: In summer, ice cream sales go up.
- Source B: In summer, more people experience wasp stings.
- "Ice cream causes wasp stings" -- states a conclusion not found in the sources.
- "Ice cream tends to correlate with wasp stings" -- reaches or suggests the same conclusion, but doesn't state it explicitly.
- "Ice cream correlates with wasp stings because both rise in summer" -- ought to be permissible IMV.
- Hope this helps explain.—S Marshall T/C 08:21, 5 June 2024 (UTC)
- Reaching a conclusion doesn't suggest a conclusion, it's a firm arrival at it. Anyway, you really aren't addressing what I had in mind regarding "reach", so let me try to express it differently. My point was based on "reach" being ambiguous, standing in equally well for "infer" and "imply". I might reach some conclusions from my reading of a text; in a text that I'm writing, I can make arguments and reach conclusions from them in my presentation to my reader. The reader may then read my arguments and reach a different conclusion from their reading of them than I had reached by the end. What they infer may be different from what I implied. Largoplazo (talk) 22:59, 5 June 2024 (UTC)
- Right. You're not responsible for what they infer, you're only responsible for what you imply. We might replace "reach or imply" with "suggest"?—S Marshall T/C 07:07, 6 June 2024 (UTC)
- Infer: deduce or conclude (something) from evidence and reasoning rather than from explicit statements.
- reach (a conclusion): to make a judgment or decision after a period of thought or research.
- Imply: indicate the existence of (something) by suggestion rather than explicit reference.
- State: express something in writing.
- Since the first two (infer and reach) are about the editor's reasoning, then we have no control over them and there is no need to mention any of them. On the other hand, the last two (imply and state) are about what they shouldn't do when adding content to an article. In other words, they are free to combine material from multiple sources to reach or infer a conclusion (for themselves) not explicitly stated by any source, but they cannot imply (hint at it) or state it (say it to our readers) when adding content to an article. M.Bitton (talk) 12:28, 6 June 2024 (UTC)
- Emphasizing what is included in the above posts, this is guidance for the writer/editor and so needs to be about actions of the writer. "State" is one of those but not really the subject of this thread. "Reach" is not explicitly an editing action (although it may lead to one). Infer is a reader action. So of the of the others, only "imply" and "suggest" meet this criteria and between those I think "imply" is better. North8000 (talk) 13:26, 6 June 2024 (UTC)
- "State" is also the subject of this thread as it's meant to replace "reach". M.Bitton (talk) 13:40, 6 June 2024 (UTC)
- Replacing "reach or imply a conclusion" with "state or imply a conclusion" seems straightforward enough to me. I think it's a good idea as it would improve clarity. Gawaon (talk) 14:01, 6 June 2024 (UTC)
- +1. The gist of my original concern was the mis-match between "reach" and "imply". I was going to fix that by changing the "imply" part, but changing the "reach" part as suggested here also works. RoySmith (talk) 14:05, 6 June 2024 (UTC)
- (ec) Of course you are right. My thinking was that "state" is a much more severe violation. I was thinking that including "state" might weaken it. Sort of like adding "don't drive at 100 mph over the speed limit" to a "don't drive over the speed limit" statement. But having "imply" in there covers it so I'd be OK with "state or imply". North8000 (talk) 14:08, 6 June 2024 (UTC)
- Replacing "reach or imply a conclusion" with "state or imply a conclusion" seems straightforward enough to me. I think it's a good idea as it would improve clarity. Gawaon (talk) 14:01, 6 June 2024 (UTC)
- "State" is also the subject of this thread as it's meant to replace "reach". M.Bitton (talk) 13:40, 6 June 2024 (UTC)
- Right. You're not responsible for what they infer, you're only responsible for what you imply. We might replace "reach or imply" with "suggest"?—S Marshall T/C 07:07, 6 June 2024 (UTC)
- Reaching a conclusion doesn't suggest a conclusion, it's a firm arrival at it. Anyway, you really aren't addressing what I had in mind regarding "reach", so let me try to express it differently. My point was based on "reach" being ambiguous, standing in equally well for "infer" and "imply". I might reach some conclusions from my reading of a text; in a text that I'm writing, I can make arguments and reach conclusions from them in my presentation to my reader. The reader may then read my arguments and reach a different conclusion from their reading of them than I had reached by the end. What they infer may be different from what I implied. Largoplazo (talk) 22:59, 5 June 2024 (UTC)
- Is there any need to have this both ways? While reading sources editors shouldn't reach or infer any conclusions not found in those sources, and while writing editors shouldn't state or imply conclusion not found in sources. I always took the sentence to mean the latter (writing), and that the former (reading) was implicit, but should it be specifically defined? -- LCU ActivelyDisinterested «@» °∆t° 16:05, 6 June 2024 (UTC)
- We don't need both. The reading part was mentioned in this discussion because of the word "reach", which once replaced, will make the sentence (about the writing) even clearer. M.Bitton (talk) 16:20, 6 June 2024 (UTC)
- We also can't control editor's minds. Reaching conclusions (in one's mind) is not and cannot be forbidden. But that shouldn't influence how the known and verifiable facts are presented to our readers. Gawaon (talk) 16:51, 6 June 2024 (UTC)
- We don't need both. The reading part was mentioned in this discussion because of the word "reach", which once replaced, will make the sentence (about the writing) even clearer. M.Bitton (talk) 16:20, 6 June 2024 (UTC)