Talk:Safety in numbers
|WikiProject Cycling||(Rated Stub-class)|
Relation to cycling
This article seems to exist mainly to support claims against helmet use on the bicycle helmets page. The article is referenced but the references are cycling helmet references, not references about 'Safety in numbers' as a concept. That makes it original research. Prospect77 04:39, 5 August 2006 (UTC)
- This article was originally created as a psychology-related stub (ref. the [2005-08-25T18:51:30] version). The material related to bicycling was added later, and was verifiable as of the [2006-07-18T07:14:07] version. While material added in the [2006-08-01T17:52:00] version lacks citations and may indeed be original research, the [2006-08-03T17:09:22] version added tags which request that citations be supplied for that material. --Wiley 13:07, 5 August 2006 (UTC)
Verifiability of sources
This document ("Review of Safety in Numbers") has been flagged for removal because it doesn't meet Wikipedia standards. Wikipedia:No original research, Wikipedia:Verifiability It is self-published at the author's web site. It has not been printed in any reliable source such as a peer-reviewed journal, textbook, magazine, book published by a respected publishing house, or mainstream newspaper. Lisamichellemurdock (talk) 17:33, 22 June 2010 (UTC)
This document ("Spurious Correlations") has been flagged for removal because it doesn't meet Wikipedia standards. Wikipedia:No original research, Wikipedia:Verifiability It is self-published at the author's web site. It has not been printed in any reliable source such as a peer-reviewed journal, textbook, magazine, book published by a respected publishing house, or mainstream newspaper. Lisamichellemurdock (talk) 05:22, 23 June 2010 (UTC)
Proposal to move content
Hi Melcombe, I've created a link here from the Bicycle safety page. The SiN page should continue to stand on its own, because Safety in Numbers encompasses more than bicycle safety. Its effects have been documented in studies of pedestrians and motorists, as well. Lisamichellemurdock (talk) 23:41, 23 June 2010 (UTC)
I've just removed a chunk of detail from the article. The reason being that it was, on the face of it, original research. The fact that there is a correlation between increased numbers of cyclists and reduced numbers of casualties per cyclist is not evidence of a causal link. For the statements to be acceptable you need an attributed and referenced claim, for each of them, that the reason for any decrease in casualties per cyclist is that there are more cyclists. -- de Facto (talk). 14:14, 29 November 2010 (UTC)
- That is an astonishing behaviour on your part, DeFacto. You have removed text supported by 41 citations, with a small army of researchers explicitly supporting the position documented in the article. It is true that a casual relationship between numbers and safety would not be established if there was data for one location only. But the evidence naturally becomes more weighty when the same relationship is found by diverse researchers at different locations around the world, and no researcher has presented a hint of what an interfering underlying influence might be. It seems that the OR has been perpetrated by your good self. I have restored your mass deletion. --Epipelagic (talk) 16:12, 29 November 2010 (UTC)
- Let's examine just the first case in the "Empirical evidence from countries and cities that have achieved substantial increases in cycling" section and see if it satisfies the policy. For Finland, the text says: "After cycling was promoted in Finland, the number of trips increased by 72% and there was a 75% drop in cyclists deaths." Who is claiming that that is evidence to support the "safety in numbers" thesis? The WP:OR policy is very clear in its requirement for statements in the article to be attributable. -- de Facto (talk). 17:23, 29 November 2010 (UTC)
I've just removed the sentence (again) stating "Substantial independent evidence supports this finding", with a string of citations, but no explanation as to who is making the claim. WP:OR is quite clear that attribution of such claims is required, and that personal speculation is not acceptable. Please don't revert without clarification of who precisely is saying it. -- de Facto (talk). 10:19, 1 December 2010 (UTC)
- I restored this removal. Even if the citations were Original Research--which they are not: read Wiki's definition--immediate removal is not the procedure prescribed. Discussion must be conducted in the discussion area before removal is undertaken. Now, DeFacto, please read the OR guidelines and explain how the 22 citations you removed meet the specified criteria. Lisamichellemurdock (talk) 01:33, 16 December 2010 (UTC)
- I've removed it again as no acceptable improvement was made when it was restored. For each one of those references we need to know whose opinion it is that it supports the hypothesis of 'safety in numbers' - with reliable references. It could be that within them they say they support the hypothesis - in which case put that in the article (with a quote perhaps), or it could be that someone else notable suggests that they support the hypothesis - in which case you need to say (with reference) who that someone is and reference where they suggest it. It remains Original Research if it is only in the opinion of the Wikipedia editor that it/they support the hypothesis. -- de Facto (talk). 20:53, 8 January 2011 (UTC)
- OK, let's just look at the first of them (Brüdea and Larsson). It isn't available free online to read, and as the abstract doesn't mention safety in numbers, I'll need to rely on your help here. Can you provide a quote from the paper supporting the assertion that it is cited to support - that "a pedestrian or cyclist, becomes less likely to be involved in a crash, as the population of similar road users increases"? -- de Facto (talk). 09:51, 9 January 2011 (UTC)
- In addition, the two articles linking from this page (Forester and Burns, FNs 24 and 25) contending that "causality" is an issue wrt Safety in Numbers are themselves Original Research. Lisamichellemurdock (talk) 01:48, 16 December 2010 (UTC)
- In Wikipedia terms "original research' means the opinion or synthesis of the editor adding it to Wikipedia. I believe that if a claim is made in a reliable source (even if that claim is untrue) it's perfectly acceptable to add it to a Wikipedia article, so long as it is correctly attributed and supported by a reliable source. -- de Facto (talk). 20:59, 8 January 2011 (UTC)
I removed the Geyer et al. reference as it doesn't seem to support the hypothesis. I don't have access to the full text, but the abstract is very explicit: "If further research shows this association to be causal, and not due to variation in intersection characteristics or selection factors involving pedestrian choice, the implications for policy and planning involving pedestrians are substantial." That implies it didn't itself establish such a link. -- de Facto (talk). 11:12, 11 January 2011 (UTC)
I've now gone through all the references that I could access looking for evidence of support for the idea. It was hard work, I don't think it is reasonable to expect readers to have to do that. I think we need to summarise the crucial point - with supporting quote in the citation perhaps - of each. I have flagged some of them as requiring a quote to aid verification. The flagged ones are tthose where the full text wasn't freely available and the abstract didn't explicitly support the idea or ones with no web url or one which wasn't in English. Please can someone (the original editor perhaps) provide the necessary quotes here please. -- de Facto (talk). 13:04, 11 January 2011 (UTC)
deFacto, you are misunderstanding the fundamental concept of verifiability. WP:V "The threshold for inclusion in Wikipedia is verifiability, not truth; that is, whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true." Each and every article that you have first removed, and now annotated, has been published by a reliable source.
I believe that you are either misunderstanding or wilfully misinterpreting the role of a Wiki editor. As is clear from the definition of verifiability, our role is not to remove information based on whether we believe it is true. Safety in Numbers is a THEORY BASED ON OBSERVABLE DATA and SUPPORTED by the cited articles that you continue to challenge. Would you visit the "Gravity" page and insist that every article supporting Newton's theory be quoted before being considered citable? The provision of the citations themselves is more than sufficient.
And finally, if you are to hold yourself out as the editor of an encyclopedia (including Wikipedia), you should most certainly show some academic rigor in the fields you are editing. Before challenging the cited articles, READ THEM. I am not required to provide you with links. It is enough that these sources exist and are available for your to acquire. And you are absolutely NOT allowed to remove them simply because you cannot or will not make an effort to do so. Lisamichellemurdock (talk) 19:53, 25 January 2011 (UTC)
- Lisamichellemurdock, let me address your points one-by-one:
- Your first point, the one about my understanding of the concept of verifiability. The second sentence mirrors exactly my explanation just a few paragraphs above; the one where I said: "I believe that if a claim is made in a reliable source (even if that claim is untrue) it's perfectly acceptable to add it to a Wikipedia article, so long as it is correctly attributed and supported by a reliable source." Yet you seem to be using it to support your assertion that I am "misunderstanding the fundamental concept of verifiability". Please explain your reasoning. -- de Facto (talk). 20:54, 25 January 2011 (UTC)
- In the point at the end of your first paragraph you appear to be suggesting that a synthesis based on data in an article that has been published by a reliable source is acceptable, even if the article doesn't make that synthesis. It isn't - that is 'original research', and covered in depth in WP:NOR. An article containing data showing a correlation between increased traffic volumes and reduced accident levels does not support the hypothesis of "safety in numbers" unless that article explicitly makes that claim; after proving a causal relationship or whatever. -- de Facto (talk). 21:05, 25 January 2011 (UTC)
- In the first point in your second paragraph you are misrepresenting my motivation for the edits I make. As I suggest in the paragraph above, reliable sources should not be used as the basis for a personal synthesis - that goes against Wiki policy - the policy that I am attempting to apply in good faith. I'm not judging the hypothesis, I'm attempting to ensure that what is written about it is supported by the references cited. I'm not challenging the articles themselves, I'm checking that they are not misrepresented or being incorrectly cited as supporting something which they don't. -- de Facto (talk). 21:15, 25 January 2011 (UTC)
- The statement in your second paragraph about requesting quotes misses the point of requesting quotes. Quotes are requested to help verify that the cited source supports the information it is purporting to support. These are particularly useful where the source isn't readily, or freely, available on line for reading and checking. -- de Facto (talk). 21:19, 25 January 2011 (UTC)
- Now the point in your last paragraph. It is unreasonable to simply assert "Substantial independent evidence supports this finding.", and then to cite umpteen, difficult to access references, without at least giving a short appraisal or summary of how each of the references support the assertion. A good article should supply all the pertinent information in the text, verifiable from the references. The reader should only have to trawl through the numerous academic papers for more detailed information such as the details of the research methodology, or whatever, not merely to establish whether and in what way they are actually supporting the blunt assertion. -- de Facto (talk). 21:47, 25 January 2011 (UTC)
- Lisamichellemurdock, it is certainly not your position that is proven, quite the contrary. All you demonstrate in that reply is; not only that you do not understand Wiki policies, but that you don't understand its ethos either. Are you refusing to cooperate in cleaning up that section of this article? Are you going to try to intimidate me each time I make an attempt at improvement? May I suggest that you read WP:Disruptive editing and WP:Assume good faith, and perhaps modify your stance on this. Our objective should be article improvement without resorting to incivility. -- de Facto (talk). 08:07, 26 January 2011 (UTC)
- deFacto, in the first instance, you removed 21 citations based on a misrepresentation of OR, and without discussion. That in itself goes far beyond incivility or a breach of what you might call the ethos of Wikipedia. Your justifications have been blatant misunderstandings or misrepresentations of Wikipedia policy. Your removals have been based on little, inconsistent, and/or circular justification. Your behavior constitutes edit warring. Lisamichellemurdock (talk) 20:43, 26 January 2011 (UTC)
- Lisamichellemurdock, my first edit was valid, as the paragraph stood was OR. Upon reverting my edit you should have addressed the issues. As you didn't it would have been perfectly valid for someone to revert your reversion as a disruptive edit. I didn't do that, I attempted to address some of the issues and point out in more detail the problems. Are you going to answer any of the points I raise and/or help improve the section? -- de Facto (talk). 22:20, 26 January 2011 (UTC)
- deFacto, no. Even if the content were OR--which it was not--immediate removal is NOT the prescribed procedure. Currently, you are engaging in circular, repetitive, and inconsistent argumentation. I will not be responding to your posts further; rather I will be reporting your edits as warring. Lisamichellemurdock (talk) 23:24, 26 January 2011 (UTC)
- A prerequisite for that would be that edit warring had occurred - which it hasn't. Have you read WP:WAR, or even WP:BOLD? Why not accept that the section needs improving, and that you can help move the quality of the article forward, rather than being stubborn about the need for change and refusing to cooperate with the necessary improvements? -- de Facto (talk). 23:38, 26 January 2011 (UTC)
of one heading to Examples of substantial increases in cycling associated with reduction in danger. The reason for this is to give a literal and accurate account of what the section says. It is a collection of primary sources which give some support to the idea of SiN. They are not - as would be ideal - secondary reviews of attempts to test the hypothesis of Safety in Numbers. Such reviews would be based on sound primary science, systematically looking at good-quality correlations of numbers with injury levels. I don't know of any such review, though I haven't looked very hard. The examples listed, so far as I can make out, are selected (important, but selected) examples. They are very low-grade evidence of the possible general rule.
As such, what I'd suggest is that this whole section should be collapsed into, perhaps, a single sentence saying that "There are many recent examples of substantial increases in cycling associated with reduction in danger" followed by the references. This strikes me as quite sufficient weight.
See Wikipedia:Citation overkill. There is a single sentence, "the risk to the individual pedestrian or bicyclist decreases where there is more people walking or bicycling.", with some 24 footnote after it, each with a lengthy quote. This is very poor style, and it suggests original research. Rather than citing a few high quality secondary sources showing there is wide agreement that something is a fact, an editor has gathered their own primary sources and assembled them to make a point. It suggests Tendentious editing, or at least that an editor was trying to win an argument on the talk page, rather than serve the needs of the reader.
Rehash of Smeed's Law
The authors don't seem to recognize that "safety in numbers" is not new and was not created for bicycle traffic. See Smeed's Law: http://en.wikipedia.org/wiki/Smeed%27s_law and Predator Satiation (cited but not referenced in the article).
Perhaps useful parts of this page could be moved to the Smeed's page?
- Smeed's Law is a general statement about safety in numbers, but the aspect related to cyclists and pedestrians is a special case, arguing that that driver behavior changes when there are enough pedestrians or cyclists present. We might want to merge Safety in numbers to Smeed's Law, but also think about a separate discussion of the special case of cyclists and pedestrians, and how that drives some opposition to mandatory helmets. --Dennis Bratland (talk) 20:14, 18 May 2013 (UTC)