Talk:Woozle effect

Original research

I thought that the reasons why I removed the Examples section and the last paragraph in the Origin and usage section were obvious but perhaps I need to explain. Let's look at the Domestic violence section. There isn't a source in the Canada subsection that mentiones the woozle effect. What someone did is take Statistics Canada stats and OCTEVAW statements and criticize them as an example of the woozle effect. The sources do not actually say "woozle". The same problem in the United Kindom subsection. Someone used official stats and statements by organizations like the Women's Aid Federation in England and made the argument that this was an example of the woozle effect. The same goes for the USA subsection where not even Gelles [1] who according to some sources coined the term uses the term woozle effect. What remains of the section Examples is the Battered child syndrome subsection and it's the only section that contains a source – Gelles & Straus (1988) – that actually says "woozle". The ref is used for only one short sentence and I won't object if someone expands and rewrites the section based on the Gelles & Straus source and writes something like "Sociologists Gelles and Straus argue that this and that is an example of..." but the rest, the Kempe study and the newspaper articles that are used as "examples", do not actually say anything about the woozle effect.

The last paragraph in the Origin and usage section contained a sentence based on a study by Gelles that didn't discuss the woozle effect and a claim that was sourced to a document by the organization "mediaradar" which appears to be a obscure advocacy group. What that document says is that VAWA advocates "misrepresent the truth of partner abuse" and that Gelles "dubs these factoids the 'woozle effect'". If that's the case then it should be possible to cite Gelles as a source and not this organization which doesn't appear to be a reliable source.

I removed the content because it was a textbook example of WP:original research and WP:Synth. There is no evil agenda as one user alleges in his edit summaries. --Sonicyouth86 (talk) 20:13, 29 April 2013 (UTC)

If you feel some of the Gelles and Straus content is retrievable then surely you should do exactly that, rather than just remove it. You seem to have a structure in mind for the content. Wikipedia relies on the expertise of all its contributors, including yours. Otherwise those less trusting than myself might think you are indulging in censorship and removing content for personally satisfying reasons.

Wikpedia should not only be fair, but also appear to be fair. Atm we have someone with expertise who is not willing to use it to improve the page. This does not look good for those browsing these talk pages.CSDarrow (talk) 16:50, 18 May 2013 (UTC)

Someone woozling won't say their woozling. Thus these are all excellent examples. Reverting censorship. — Preceding unsigned comment added by 64.7.147.17 (talk) 20:31, 14 May 2013 (UTC)

Someone woozling won't say they're woozling, but what we need is reliable 2nd and 3rd party sources saying that they're woozling, not some anonymous random person on the internet saying they're woozling. Please have a look at the policies on original research, original synthesis and reliable sources that Sonicyouth86 linked above. The allegations of censorship or some evil agenda are pretty disingenuous. Cheers, Dawn Bard (talk) 19:57, 16 May 2013 (UTC)

If something is unequivocally and self evidently woozling there would be no need for what you suggest. The common sense and good judgement of the editors can be used. Accusing others of being disingenuous is probably not a good thing. CSDarrow (talk) 18:53, 18 May 2013 (UTC)

There would be no need for what I suggest? All I suggested was that reliable 2nd were required, which is completely standard at Wikipedia, as I'm sure you know. Dawn Bard (talk) 19:07, 18 May 2013 (UTC)

even if there were a third party source that says "So and so woozled"- so what? how is that encyclopedic? there would need to be some additional commentary /analysis showing how or why the woozling was an important or relevant example. -- TRPoD aka The Red Pen of Doom 19:25, 18 May 2013 (UTC)

Now that would be an interesting precedent to set. Very few of the citations on gender issues do more than just state an undemonstrated position. I would join you in support of that. CSDarrow (talk) 02:57, 19 May 2013 (UTC)

KEEP THE EXAMPLES

After a thorough read of WP:OR and WP:SYNTH, it is my conclusion that the examples that were provided should stand (and be restored).

"SYNTH is original research by synthesis, not synthesis per se." Consequently, the examples are OR-free.

1 : The examples provided are just that. Mayhaps someone would prefer the title Illustrations? Either way, the examples are there to demonstrate real-life woozles, as properly-defined earlier in the article. To my reading, they provide zero new, or original research. There is no pretense of adding new information to the well-supported definition and description of Woozle.

2 : The examples are legitimate statements of fact. Criticism involves opinion and is an inappropriate description of the contents of these examples. If someone wants to verify the sources of each example to ascertain their authenticity, go ahead, and remember to provide a detailed comment of your findings (preferably in TALK, first). This gives the author (and other editors) the means to assess your findings and, where applicable, correct their work. It is also professional and courteous (IMO).

3 : Take, for example, the second illustration within Domestic Violence - Canada. The "woozley" article cleary cites Statistics Canada 2011. Ah!, you say, yet the StatsCan info being compared to is from 2009. Yes it is ; and if you look up the 2011 StatsCan document referred to in the woozle, it in turn references the 2009 data. Conclusion : this is a bona fida example of woozle. The conclusion was NOT arrived at via WP:SYNTH (and even less so via WP:OR). It was arrived at via synthesis.

4 : For heaven's sake, let us all remember that the intent of WP policy is to limit to a minimum leading or misleading entries, especially the grossly so. WP policy is NOT intended to be legalistically applied universally at the minutest levels of detail everywhere. Because we are human and are wont to get carried away in our enthusiasms, pages such as WP:What_SYNTH_is_not were created to remind us.

There has been a lot of back and forth on this article, so I am posting this assessment in TALK before proceeding with restoring the examples. If you disagree with restoring the examples, please refute clearly points 1, 2 and 3. -- Visispace (talk) 05:20, 4 October 2013 (UTC)

WP:NOR says "To demonstrate that you are not adding OR, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented". Until that is done, the examples should not be in the article. Dougweller (talk) 12:46, 4 October 2013 (UTC)

Notability

Help requested. This article may not have (WP:Notability) significant coverage from reliable, independent sources.

I searched for sources. First I searched Google (https://www.google.com/search?q="woozle+effect"&start=990 less than 300 hits) but the sources were not reliable. Nearly all were small, inactive blogs and self published (WP:SPS). An exception, AVoiceForMen, lacks editorial integrity: Southern Poverty Law Center keeps the site on a watchlist of "woman-hating sites" and notes it "is essentially a mouthpiece for its editor, Paul Elam".

I searched again, excluding many domains (wordpress.com, avoiceformen.com, reddit.com) to avoid WP:SPS. Hits are then mostly primary sources. That's not ideal: WP:Notability "secondary sources [...] provide the most objective evidence of notability" and '"Significant coverage" addresses the topic directly and in detail, so that no original research is needed'.

For primary sources, Google Scholar has 74 hits for "woozle effect" which is small for a body of research. Of the first ten hits, 4 are written by RJ Gelles, and at least another 4 cite Gelles.

From the current references, Donald D. Dutton does seem to be a good source: at least his book "Rethinking Domestic Violence" covers the topic significantly. He is 6 of the 24 references, but for notability "Multiple publications from the same author or organization are usually regarded as a single source for the purposes of establishing notability." The rest appears to be primary sources all, when we'd prefer secondary, and some of it does not mention "woozle effect" but instead only uses "woozle" (the same way it uses Heffalump, as a figment or something elusive)--parts of the Origin and Usage section may be misleading for the same reason.

My tentative evaluation: "Woozle Effect" may be used by a very small number of academics but may not yet reached the Notability for a standalone page. One solution may be to create a standalone for Woozles which mentions Woozle Effects. That is the treatment given to "Heffalump" and "Heffalump Trap" even though Heffalump Trap seems more notable than Woozle Effect.

Yamrway (talk) 03:22, 4 February 2014 (UTC)

I don't know, I think there's definitely enough discussion of it in reliable sources to establish notability. Try google scholar instead of google to get higher quality sources: [2]. Also, what in the world could you mean by "primary sources" in the context of a concept in social science?— alf laylah wa laylah (talk) 03:29, 4 February 2014 (UTC)

I did search Google Scholar (read above...). Social sciences have primary sources too, otherwise it wouldn't make sense to direct me to search for research.

I do not see enough "discussion." Outside of a very small number of academics using the phrase incredibly sparsely, there's no discussion. No secondary sources report on this use. We could likewise have a standalone page for every single time "Heffalump Trap" was used (and there's far more notability there, honestly) but wiki is not for lists like that unless the topic is Notable. Yamrway (talk) 03:35, 4 February 2014 (UTC)

How are you using the term "primary sources" here, then? Also, if you accept Dutton, why not Gelles, who discusses the term at reasonable length in the sources cited in the article, e.g. Violence in the Family:

The "Woozle Effect" begins when one investigator reports a finding, such as Gelles's (1974) report that ... [blah blah blah] ... In the "Woozle Effect," a second investigator will then cite the first study's data, but without the qualifications (such as done by Straus, 1974a). Others will then cite both reports and the qualified data gain the status of generalizable "truth."

There's more along these lines in Gelles's other work. Makes two sources, which is usually enough to meet GNG.— alf laylah wa laylah (talk) 03:50, 4 February 2014 (UTC)

Mistaken references

There appears to be a striking divide between the referenced sources and the claims made about those sources on this page. Editors beware, may need many NPOV changes.

Roth, Philip L.; SWITZER III, FRED S. (1999).

Roth is not discussing the Woozle Effect at all. He is discussing approximation techniques called "missing data techniques." Missing data which is to be approximated is referred to a Woozle (or Heffalump), the only context the word is used. The article does not mention cognitive or social bias, or repeated research citing at all. The sentence which cited Roth:

Selection of data and design of research instruments to gather raw data are linked to the creation of the Woozle effect on many fields of study.

is in violation of WP:NOR (or needs a new citation) because that is not Roth's conclusion or even on his topic or field. This sentence is also similar to a later sentence ("...the presence of the woozle effect in many areas..."). I have removed the sentence for now to reduce redundancy and the need for a new citation.

Roy F. Baumeister; Kathleen D. Vohs (2007).

The full text of this volume is searchable online. It does not contain the word "woozle." The sentence which references it,

The woozle effect is seen as an example of confirmation bias[16] and linked to belief perseverance.[17]

is sufficiently sourced already because [16] "Rethinking Domestic Violence" discusses woozles, confirmation bias, and belief perseverance. I have moved reference [16] to the end of this sentence and removed this reference.

Donald G. Dutton. (2006)

4 redundant references, switched to {{rp}} format.

Kinchin, Niamh (20 MAR 2007).

The full text of this is searchable online (quick google search). "Woozle" does not appear in the body, only in the title of a cited work. I have removed the reference and edited the sentence citing it accordingly.

Kinchin, Niamh (20 MAR 2007)

Again this reference uses "Woozle" in a way wholly unrelated to this article:

The reader is obviously familiar with Wohlwill's (1963) infamous woozle hunt. It would appear that the woozle, that is, a general agreement concerning the underlying processes in human cognitive development, remains as elusive as ever.

For starters, this usage in 1963 predates the coining of "Woozle effect" as described in Wiki's article (1979). More, Kinchin's use is directly opposed to "Woozle Effect." Woozle Effect is a problem of cognitive or social bias, but Kinchin's Woozle is a desirable, elusive, and accurate consensus: a synonym for "prize." I have removed the reference and edited the sentence citing it accordingly.

Many remaining sources are paywalled and I urge editors to try to check their accuracy (as I will) because so many citations here are in error.

Yamrway (talk) 05:32, 4 February 2014 (UTC)