Wikipedia talk:Conflicting sources

BLP, and other comments
It might be a good idea to mention that other policies may dictate which sources to use in certain circumstances. WP:BLP comes to mind. We are certainly not going to include both sources for a BLP if one reports something negative about a living subject that is conflicted by an equally reliable source. In general, I'm also wondering if the ideas of this essay aren't already understood by way of our other policies. 1) When one source is more reliable we already defer to that source. 2) When they are equally reliable, and there are no other policy issues like BLP, we already report both per WP:NPOV. I guess it might be helpful to clearly state this in one essay, but we should ask ourselves if this essay is needed. Cheers.Griswaldo (talk) 16:25, 21 February 2011 (UTC)


 * Whether an essay is really "needed" is often questionable. But since it's covered explicitly in no policy or guideline as far as I can determine, it could serve as a good place to refer people to when they ask that same question. Sure, the answer can be deducted from existing policies and guidelines but when an user asks you what to do, it's much nicer if you could just link them to a place where his question has been answered already.
 * And the BLP-thing you mention actually proves that point. Where in WP:BLP does it say that we cannot add information of conflicting sources that are both reliable? In fact, it says that such articles have to adhere strictly to NPOV, V and NOR and NPOV, as you, actually requires us to report both. Regards  So Why  16:41, 21 February 2011 (UTC)
 * It does not require us to report both at all. Where does it say that?  What is says about contentious material is this (with emphasis added to illustrate my point):
 * "Wikipedia's sourcing policy, Verifiability, says that all quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation; material not complying with this may be removed. This policy extends that principle, adding that contentious material about living persons that is unsourced or poorly sourced should be removed immediately and without discussion. This applies whether the material is negative, positive, neutral, or just questionable, and whether it is in a biography or in some other article."
 * Any contentious piece of information in a BLP with two contradictory sources only, is by its nature "poorly sourced". We assume that publications usually deemed reliable are reliable, but we also know that they aren't perfect.  When two such publications contradict each other one of them has clearly made a mistake.  If they are being used to source contentious material BLP mandates that we don't use either.  We look for better sources, or additional sources to sort this out.Griswaldo (talk) 17:04, 21 February 2011 (UTC)

MEDRS--one vs many
The current mess going on over at Weston Price demonstrates the need for this guideline. Currently ONE source is being used to make what is at best misleading statements about Weston Price and focal infection theory even though there are literally dozens that say otherwise (see Reliable_sources/Noticeboard for the details of just WHY this source is so wrong) When a medical textbook can be PROVEN to be factually wrong using other medical textbooks and journal it should NOT be allowed no matter who publishes it.--BruceGrubb (talk) 19:05, 22 February 2011 (UTC)
 * as the discussion at the RSN clearly shows, the textbook cited represents the major stream of accepted medical thought. It certainly cannot be proven factually wrong. The claim that it is is in contradiction to the overwhelming preponderance of modern evidence. As the standard recent work on the field, it's the ideal sort of summarizing source we look for. The evidence otherwise is a few equivocal statements saying in the usual polite academic language that the minority theory cannot altogether be ruled out. There is no current professional RS that actually supports the theory. This is really exceptional: an attempt to defend what is now a small minority viewpoint not by the usual claim that one source supports it, but by the claim that almost all sources support it!.  There is no point in further discussion here. There is no conflict of sources.    DGG ( talk ) 01:35, 24 February 2011 (UTC)
 * Actually as shown by my response the RSN does not address any of the problems in this one source (Ingels).
 * "In the 1930s, editorials and research refuted the theory of focal infection..." ) Ingles Textbook 2002; 2007)
 * "Today the concept of focal infection has been integrated into the practice of medicine. One speaks no longer of the theory of focal infection; one recognizes focal infection as a definite pathologic condition requiring scientific diagnosis and treatment." (1947 Journal of the American Medical Association Volume 133:2 page 111). This statement was repeated word for word in Review of gastroenterology Volume 18 pg 71 of the National Gastroenterological Association in 1951.


 * "The factor of focal infection still seems to be a most important one in determining the type of treatment of infected tonsils." (1957 Journal of the American Medical Association: Volume 163, Issues 5-8; pg 519)


 * "This introduction to the focal infection of disease should not be construed to mean that the theory has no basis in fact. There is little doubt that under certain circumstances microorganisms can move from one area of the body to another to establish their customary pathology in another locale. (Pallasch, Thomas J. DDS, MS; Michael J. Wahl, DDS (2000) "The Focal Infection Theory: Appraisal and Reappraisal" Journal of the California Dental Association)


 * The following sources also show the Ingels textbook to be factual wrong regarding FIT either in history or definition:


 * Bergenholtz, Gunnar Preben; Hørsted-Bindslev, Claes Reit putlich (2009) Textbook of Endodontology Wiley page 135-13


 * Ghom (2009) Textbook of Oral Pathology Jaypee Brothers Medical Publishers pg 459


 * Henderson, Brian; Michael Curtis, Robert Seymour (2009) "Periodontal Medicine and Systems Biology", Wiley; Page 33


 * O'Reilly, PR Claffey NW "A history of Oral Sepsis as a cause of disease" Periodontal 2000 1997


 * Pallasch, Thomas J. DDS; MS, and Michael J. Wahl, (2003) "Focal infection: new age or ancient history?" Endodontic Topics, 4, 32–45 Blackwell munksgraard


 * Saraf (2006) Textbook of Oral Pathology Jaypee Brothers Medical Publishers pg 188


 * Note that that three of these are also textbooks. Furthermore the "(i)t certainly cannot be proven factually wrong" claim is nonsense as the above shows.  Finding a 1957 article by no less then the American Medical Association that says the "factor of focal infection still seems to be a most important one" proves the "In the 1930s, editorials and research refuted the theory of focal infection..." ) to be factually wrong does it not?  One source cannot be used to over rule other that prove that the source is talking nonsense.--BruceGrubb (talk) 05:10, 2 March 2011 (UTC)

Primary material vs third party reviews of older primary material
One problem we are having in the Weston Price and Focal infection theory articles is that the anti-root canal group is using his 1923 work and this is what all the counter material focus their attention on. However Price himself makes comments in 1925 and 1939 reliable sources that put a different light on his views but because these are primary sources they have been kept out of the respective articles. Shouldn't WP:NPOV allow such material if everybody is missing it?--BruceGrubb (talk) 22:14, 8 May 2011 (UTC)

How (not) to handle conflicting sources
The section Handling conflicting sources tells us, "In many cases, when two (or more) reliable sources conflict, one (or more) of those sources can be demonstrated to be unreliable." How can reliable sources be demonstrated to be unreliable? I a source can be demonstrated to be unreliable, it is not a reliable source. I see at least two possible interpretations that would remove the internal inconsistency. The first is that "reliable" means "seemingly reliable" – at a first glance the source seemed to be reliable, but a further examination showed it to be a self-published book by someone who is not a recognized expert on the topic. The second interpretation is that "reliable sources" refers to the specific Wikipedia terminology as set forth in Verifiability and Identifying reliable sources, whereas the term "unreliable" should be interpreted in the common dictionary sense of "not dependable in accuracy". These interpretations are different, and the text should be made less ambiguous while removing the apparent internal inconsistency.

In any case, the underlying idea is that sometimes a conflict can be resolved by determining that a conflicting source is unreliable and then discarding it. But then, the following section, How not to handle conflicting sources, directs us editors: "Do not choose which one is 'true' and discard the others as incorrect." This would seem to contradict the remedy offered by the first section. --Lambiam 20:24, 18 February 2018 (UTC)

Summary sources lacking detail
This essay suggests that editors "Prefer up-to-date sources." While I can understand this to be a reasonable thing to do in an academic context where evolving research brings new thinking and opinion to light as researchers dig deeper into a particular topic. But is this a reliable guideline when the more recent sources while summarizing the older ones also conflate or misinterpret the historical record? In historical situations, the older sources are closer in time to a particular events and will often report upon these events in far more detail, and are probably more factually accurate. One can see this happening in Wikipedia, where an article might incorrectly interpret a source or make statements that are not supported by the cited source(s), or do not cite sources at all. I would also suggest that editors should consider the following points when faced with conflicting sources: - Cameron Dewe (talk) 00:27, 27 November 2022 (UTC)
 * "Prefer the source(s) that a source relies upon." - If a source relies primarily on another source, then check that source too, and use it. Sometimes multiple sources all rely on a common single source for their facts. This especially applies with news stories that are syndicated by multiple news outlets. Often a news story is credited to a particular news agency, yet is slightly rewritten for editorial reasons, perhaps rewording a headline, summarizing or shortening an article, rearranging it or supplementing it with additional reporting or opinion. One can look at a range of outlets and see how the same article changes in different publications. Often, it is possible to access the agency's original report, or see early publications of a report that are scoops or hot off the press, compared to others that have taken these earlier reports and reprinted them. Sometimes, there are typos or transcription errors introduced by this process. This means the sources closer to the original report are likely to be less error prone.
 * "Prefer sources that provide eyewitness details." - Sources that provide multiple detailed accounts by various eyewitnesses and explains or resolves conflicting accounts are preferable to one that just provides a summary of those accounts. A source that depends on a single eyewitness account might miss details that other eyewitnesses report. A source that has to balance conflicting reports from different eyewitnesses, and successfully addresses that balance, is likely to be a less biased account than one that accepts the (unchallenged) view of a single person.
 * "Prefer sources that can be verified from their own cited sources" - Sources that are transparent about where they obtained their information from are preferable to those that do not cite a source, especially where the facts are in question, and sources summarize those facts.