Talk:Neuro-linguistic programming

corrections: Satir, Perls

The article currently says that Bandler and Grinder [modeled] "Virginia Satir so they could produce what they termed the Meta-Model". This is not consistent with what Bandler and Grinder claimed, and is at odds with the scholarly sources. Virginia Satir was the co-founder of the Mental Research Institute and was a prominent member of the Palo Alto Group. (see Wilder, 1979) as cited in [1] and they later co-authored a book together with her titled "Changing with families" (ISBN 0-8314-0051-X) published by Science and Behavior Books in 1976. However, according to Bandler and Grinder, the Meta-Model was modelled from Fritz Perls (Gestalt therapy) in addition to Virginia Satir, and the Milton-Model was "modeled" from Milton Erickson. Amit Pinchevski (2022)[2] citing Dr Robert S. Spitzer, a Prominent American Psychiatrist and former Chair of American Psychiatric Association's DSM taskforce, states that "In fact, NLP was conceived after Bandler (a university student at the time) was hired to record and transcribe both a month long workshop that Satir conducted in Canada and teaching films featuring the psychiatrist Fritz Perls. Bandler spent several months listening meticulously to the recordings, eventually adopting many of the two’s voice patterns and mannerisms" (see Spitzer, 1992). Karen Stollznow who is also cited in the same paragraph in her book also acknowledges that Meta-Model was modeled from Perls and Milton-Model from Erickson (See p.236 of Stollznow 2014 book 'Language Myths, Mysteries and Magic). Stollznow asserts that these models do not accurately represent those individuals. --Notgain (talk) 06:16, 3 May 2024 (UTC)

Just wanted to make a correction. Dr Robert S. Spitzer (the one who published Bandler and Grinder's first book The Structure of Magic) is a neurologist but not the renowned Robert L. Spitzer who chaired DSM-III taskforce and was probably the most well-known psychiatrist of his time as I mentioned earlier. --Notgain (talk) 08:29, 5 May 2024 (UTC)

Shortened Citation Notes

The article currently uses a mix of referencing styles and there are missing page numbers for quotes or what may or may not be paraphrased text but we don't know because there are missing page numbers. See Template:Sfn for a guide how add page numbers and quotes. --Notgain (talk) 01:34, 6 May 2024 (UTC)

• Oppose, it's disruptive. See WP:REFVAR, which requires a WP:CONSENSUS from the regular editors of the page before you may do so: "Editors should not attempt to change an article's established citation style, merely on the grounds of personal preference or to make it match other articles, without first seeking consensus for the change." Skyerise (talk) 23:25, 6 May 2024 (UTC)

  I would contend that it was not "merely on the grounds of personal preference". I was looking at the best practises in other Good and Featured Article candidates. I'm personally most comfortable with the APA format but done research papers using Harvard referencing style with footnotes. I was thinking that style was the best for this article. Given that the article covers critiques from counseling psychology, coaching psychology, communications theorists, sociology and linguistics, its not simple. Do you have examples of article with similar content and multidisciplinary critiques? What referencing style worked best? --Notgain (talk) 06:18, 7 May 2024 (UTC)

I also notice that @Newimpartial: reverted you on several occasions just after you broke a bunch of citations by trying the same thing back on 5 May, a day or so before I noticed what you were doing. So that's 2 opposed. Skyerise (talk) 23:41, 6 May 2024 (UTC)

I’ve already explained this and it is also in the edit comments. I fixed that citation errors. I didn’t know about the display error setting which was off by default. Again, I did appreciate your help. When I get more time, I’ll go back and justify each change. —Notgain (talk) 03:30, 7 May 2024 (UTC)

• Oppose, as Skyerise explained to you, you can't just change citation style without consensus. If you want to do changes you have to clearly justify them in order to show other editors your reasons or concerns about it, and if these go according to the WP:CS. Rodrigo IB (talk) 04:52, 7 May 2024 (UTC)

You're absolutely right about needing consensus. My first step should have been proposing these referencing changes here on the talk page. Would you be willing to join a discussion about how to best improve the consistency and verifiability of the article's references? --Notgain (talk) 06:01, 7 May 2024 (UTC)

As the WP:CS sustains. Citations are key for verifiability. Looking at the changes you did, im concerned that these could compromise the access of common editors and readers to those sources. Which is very crucial for this article.

Editors with their own personal bias can incur in practices (like meat-puppetry) that violate WP:V,WP:NPOV,WK:STYLE.

The controversies sorrounding NLP obligate us as editors to make sure we are not doing original research. Which, for surprise of no one, has to be verified by others. For that reason, i think is naive to compare it to other articles just because different citation styles were used, or due to their extensivity in other disciplines. Rodrigo IB (talk) 06:29, 9 May 2024 (UTC)

• Oppose There are several ways to add pagenumbers in/with reftag-refs (not surprisingly), including Template:Rp. Gråbergs Gråa Sång (talk) 05:46, 7 May 2024 (UTC)

  The inconsistent styles and missing page numbers make WP:V and editing difficult. Have you come across similarly complex articles that successfully used {{Rp}} or other templates to maintain readability while ensuring accurate citation information? Especially ones covering multiple disciplines, as this article does? The immediate issue is that there are paraphrasing of sources without clear page numbers which makes WP:V difficult. Another issue is that are duplicates of the same sources across the article. That was an advantage of using {{efn}} and {{sfn}}. We are already using {{r}} in the article. <ref> is also often combined with {{sfn}}. Also some of the quotes in the current article are inside the cite element when they would be better handled as an {{efn}}. We have critiques from linguistics, counseling psychology, anthropology and sociology. --Notgain (talk) 06:12, 7 May 2024 (UTC)

  I have nothing against sfn etc as a style, though ref-tag is always my choice when I start articles, with rp if necessary. IMO reftag is generally more understandable for general and new users, and both VE and source editors benefits from named refs if used. But an article should be consistent, and if consensus here is to use sfn or whatever, that's fine. Gråbergs Gråa Sång (talk) 06:21, 7 May 2024 (UTC)

  Fwiw, my knee-jerk reaction when scrolling through the ref section, is that "traditional reftag" seems to be the majority use, so if I was to start working on consistency, I would change the "Jeremiah 1995." style ones and get rid of the "Works cited" sections. But if the primary/secondary division is considered valuable, that might not work. I think some Wikipedians consider the more academic look of sfn-style a mark of quality. Gråbergs Gråa Sång (talk) 06:31, 7 May 2024 (UTC)

  I'm told that {{cite Q}} would solve one of the issues I had with existing use of <ref name="Joe-1995">{{cite journal|...}}</ref>. {{cite Q}} enables you to pull the reference data from wikidata by using its Q ID. It was too verbose and made it difficult to maintain especially in source mode. My proposal is for any citations that are current citations that are defined inline such as "<ref name="Joe-1995">{{cite journal..." that if that citation is on wikidata then we is replace it with "<ref name="Joe-1995">{{cite Q|...". That will reduce some of the clutter and retain existing r and rp template use. Then we can use r and rp. Then if there is consensus to use sfn then we can adopt that together with efn which is already in use in the current article. --Notgain (talk) 08:32, 7 May 2024 (UTC)

  No, you should revert to the previous citation style per consensus and WP:REFVAR, full stop. Remsense诉 09:20, 7 May 2024 (UTC)

  Do you know what I mean by <ref name="...">{{Cite Q|...}}? It just moves the clutter of the reference out of the content. That is one of the biggest issues with the article in its current state. Its still using the same citation style. It is a wrapper for {{Citation}} that returns formatted citation from statements stored on a Wikidata item (referred to by its Q identifier or QID) for citable source. It would be a good interim solution while consensus is sought for sfn which is my preference as it would be far more professional. efn has been used in the article for years. --Notgain (talk) 10:19, 7 May 2024 (UTC)

  Yes, I know what it means, I've used it myself. It is a different method of referencing, even if the output looks the same. The words "method" and "style" are used interchangeably on the guideline page, but the reason underlying changes in both is the changes are disruptive to others, hence why the guideline is to defer to the first format used in a dispute: other editors who want to edit this page don't want to suddenly swap to having to look up Wikidata codes. You seem increasingly unwilling to understand that. Remsense诉 10:33, 7 May 2024 (UTC)

  Do you understand that we are currently using <ref>{{Cite Journal |...}}</ref> in the article and post people use tools already like [3] to populate the details of that from the DOI, ISBN, etc. So using <ref>{{Cite Q |...}}</ref> might actually be less work, and they'd be familiar anyway. The editors who don't undertstand wiki syntax usually use a visual editor or they just rely on other wikipedians to clean up after them. I guess we'll need to wait for others to chime in with their preferences. --Notgain (talk) 13:13, 7 May 2024 (UTC)

  Do you understand that whether you think they'd like it better doesn't matter? Remsense诉 13:17, 7 May 2024 (UTC)

  some like the simplicity of ref even thought sfn is technically better. There are featured articles that use ref only but the longer ones with notes and many references prefer sfn. —Notgain (talk) 04:45, 8 May 2024 (UTC)

  Again, as pointed out just above, that is irrelevant here. The only relevant thing is whether you have consensus or not. Clearly, you don't. Skyerise (talk) 11:26, 8 May 2024 (UTC)

  I’d like to hear the arguments for and against sfn v ref, with examples. Besides none of the regular contributors to this specific article have raised objections so there is no evidence of clear consensus from regular editors. I have enabled errors so I can correct the errors you complained about. I think now consensus can be sought through editing and discussion. —-Notgain (talk) 21:11, 8 May 2024 (UTC)

  No, the burden to demonstrate a change is preferable is on you. If no one agrees, then you may not make the change. (You have; no one has; you may not.) People are entitled to establish consensus regardless of contributions; frequent editors do not own the articles in question. Remsense诉 06:43, 9 May 2024 (UTC)

  I must have misunderstood what you were saying on AN/I. I thought your earlier point was that changing from ref to sfn referencing format would be unwelcome because of the learning curve for the existing or previous editors of this article, or that existing editors might not like it. You said, "the changes are disruptive to others" (above). I assumed you were referring to previous editors of this article. How could it possibly be disruptive to edits who have never edited this article? I assumed you meant you needed to obtain consensus from them (previous editors of this article). None of them have commented yet. However, the silence from the previous contributors could be interpreted in different ways. It could mean that they are indifferent to the change, that they are unaware of the discussion (most likely scenario), or that they are still forming opinions. Anyway, I'm going to help out at Category:Harv and Sfn no-target errors not to recruit or canvas support but to learn more about the interaction between sfn, efn and ref formats - as well as learning more about WP:V --Notgain (talk) 16:30, 9 May 2024 (UTC)

  I'd note that converting references to Cite Q en masse would be contentious even without CITEVAR. -- LCU ActivelyDisinterested «@» °∆t° 15:54, 7 May 2024 (UTC)

  my intention for suggesting sfn was to enhance readability and maintainability. With sfn, you define the reference using cite templating in the bibliography. Assuming ref is inadequate too, do you know of an alternative solution that meets that need given the huge number of citations on this article? —Notgain (talk) 02:10, 8 May 2024 (UTC)

  120 inline citations is simply not a particularly high number, and is adequately accommodated by any common means of citation. Remsense诉 06:48, 9 May 2024 (UTC)

  I'm well aware of how {{sfn}} templates work, but your supposition that ref tags are inadequate is simply your own personal opinion. You won't find any concensus that one form of referencing is better than another, the editing community is deeply split on the matter. This is why CITEVAR warns against changing style types, as it causes unnecessary drama that wastes editors time. -- LCU ActivelyDisinterested «@» °∆t° 10:55, 10 May 2024 (UTC)

  Afaik, neither VE or WP:REFTOOLBAR has any "format ref as sfn/harvn" option, is that correct? Also, no ref-tag, no named ref. Gråbergs Gråa Sång (talk) 12:15, 10 May 2024 (UTC)

  Kind of, VE supports adding templates so it supports {{sfn}} (as long as you know what they are), the same would be true of {{r}}, {{ref}}, {{efn}}, etc. I don't think the REFTOOLBAR point is relevant, if you already using source editing then using the toolbar to format sfn/harv would take longer than typing it.
  I don't think REFTOOLBAR has any ability to re-use a refname, but again it would be quicker to type it, VE certainly can though. -- LCU ActivelyDisinterested «@» °∆t° 10:46, 11 May 2024 (UTC)

  Reftoolbar absolutely has the ability to re-use a refname, "Named references", to the right of the ref-template drop-down. Wikipedia:RefToolbar/2.0. In VE it's Cite > Re-use. In source, you name them with the "Ref name" field in the template window. Gråbergs Gråa Sång (talk) 10:56, 11 May 2024 (UTC)

  Your right, I had missed that in REFTOOLBAR. -- LCU ActivelyDisinterested «@» °∆t° 12:05, 11 May 2024 (UTC)

• Neutral. I'm not super educated on the nuances of citation styles, but I feel like the citation style used on this article in particular is not super important. I think the bigger issue is that when Notgain tried to convert it all to {{sfn}} without gaining consensus, they did so incorrectly, and broke citations in the process. I've used {{sfn}} and tend to prefer it with more complicated articles such as this one, but if other editors are opposed, I'm prepared to respect that; I'm not convinced Notgain is, which is another issue. (Also, the corresponding ANI thread on this issue ended without clear consensus and without admin closure; I'm not sure what to make of it, but it feels relevant.) 〜 Askarion ✉ 16:41, 13 May 2024 (UTC)

  sfn was it is easier to read in source mode but I now have source highlighting so I’ve settled. I’m not going to push sfn on the great unwashed. —15:38, 14 May 2024 (UTC) Notgain (talk) 15:38, 14 May 2024 (UTC)

Druckman & Swets 1988

Before I forget, can someone who has editing rights update the reference to Druckman & Swets 1988 report? The consensus of the committee was discussed in chapter 8. Note that the DOI in the current reference to Druckman&Swets 1988 is incorrect (it points to a book review of the committee's consensus report, not the report itself), please change to: {{cite book | title=Enhancing Human Performance: Issues, Theories, and Techniques | chapter=8: Social Processes | pages=133-166 | publisher=National Academies Press | publication-place=Washington, D.C. | date=1988-01-01 | isbn=978-0-309-03792-1 | doi=10.17226/1025 | ref={{sfnref | National Academies Press | 1988}}}} or if you want to include the editors: {{cite book | last1=Druckman | first1=Daniel | last2=Swets | first2=John A. | title=Enhancing Human Performance: Issues, Theories, and Techniques | chapter=8: Social Processes | pages=133-166 | publisher=National Academies Press | publication-place=Washington, D.C. | date=1988-01-01 | isbn=978-0-309-03792-1 | doi=10.17226/1025 | ref={{sfnref | National Academies Press | 1988}}}} That was a honeytrap for some researchers copy and pasting from wikipedia without checking sources. Otherwise, there's the named reference version for those who prefer that style: <ref name="Druckman-1988">{{cite book | last1=Druckman | first1=D. | last2=Swets | first2=J. | title=Enhancing Human Performance: Issues, Theories, and Techniques | publisher=National Academies Press | publication-place=Washington, D.C. | date=1988-01-01 | isbn=978-0-309-03792-1 | doi=10.17226/1025 | pages=133-166 | chapter=8: Social Processes}}</ref> --Notgain (talk) 04:13, 10 May 2024 (UTC)

The use of Druckman and Swets (1988) as a reference to support the statements #1 "Numerous literature reviews and meta-analyses have failed to show evidence for NLP's assumptions or effectiveness as a therapeutic method" and #2 "Bandler led several unsuccessful efforts to exclude other parties from using NLP" is problematic. Druckman (2004) clarifies that the panel evaluated techniques like NLP for their potential in "enhancing learning, improving motor skills, altering mental states, managing stress, or improving social processes." The panel's focus was on NLP's potential for social influence, not its therapeutic applications. They found NLP's assumptions and effectiveness in social influence to be unsupported by psychological evidence. Its worth noting that the panel was "impressed with the modeling approach used to develop the technique," this interest in modeling does not directly speak to NLP's effectiveness as a therapeutic method. The fact that the planned NLP training was not implemented could suggest the type of "unsuccessful efforts" hinted at in statement 2, but this remains speculative. I couldn't find anything in the cited source to directly support statement 2. Therefore, it's recommended to remove Druckman and Swets (1988) as a supporting reference for these two statements. --Notgain (talk) 08:23, 10 May 2024 (UTC)

•Denied, while the Druckman and Swets (1988) aim is not the therapeutic effectiveness of NLP, it touches the lack of empirical evidence on representational systems, you even quoted this from the article: "Numerous literature reviews and meta-analyses have failed to show evidence for NLP's assumptions OR effectiveness as a therapeutic method"

The review is clearly relevant. Rodrigo IB (talk) 18:40, 10 May 2024 (UTC)

I think there may have been a misunderstanding here. Statement #1 was from the current article, not the source. The NRC (Druckman and Swets 1988) did not review NLP as for its therapeutic application. And you have have not addressed statement #2 which is not suppprted by the source either. If you think it is please provide page numbers to substantiate for verifiability. —Notgain (talk) 20:57, 10 May 2024 (UTC)

It seems you're reading statement #1 as "and/or", which would make the Druckman & Swets (1988) source relevant because it addresses the lack of empirical evidence for NLP's assumptions. However, if we interpret "or" to mean that both parts of the statement need separate supporting evidence, then a source that doesn't address NLP's therapeutic effectiveness might not be suitable for this statement. It is important to distinguish between NLP's assumptions, and its effectiveness in different areas of application - whether it be therapeutic, management or social influence, as we discussed earlier. To be clear while the NRC (Druckman & Swets 1988) provides a strong review into NLP's assumptions, it does not directly address its therapeutic effectiveness. Other reviews do. Therefore, I’d prefer to cite separate, relevant sources for each part of statement #1. This will aid in WP:V. —Notgain (talk) 21:40, 10 May 2024 (UTC)

What? Where is stated that the sources for that particular case should adress both?

Even you proposed a section around persuasion, which is one of the different approaches of NLP. The whole article, including that single sentence is referring to NLP in general.

It gets worst when we analize your own statement: "However, if we interpret "or" to mean that both parts of the statement need separate supporting evidence, then a source that doesn't address NLP's therapeutic effectiveness might not be suitable for this statement."

For your own argument then a source that adresses just one aspect is still valid, because it's providing evidence for a specific claim; it would be a problem if and only if was the only source cited to sustain the lack of evidence in regards to the therapeutic approach of NLP; which is not the case.

The "interpretation" (which this is not about) you highlight plays against you.

I don't get it. Rodrigo IB (talk) 22:08, 10 May 2024 (UTC)

I hear your point about the use of the word “or” in the statement #1 and how it could be interpreted to mean that a source addressing just one aspect is still valid. However, my premise is that for a more accurate representation of the sources, it would be ideal if each part of the statement is supported by citing relevant sources that directly address the respective claim in line with WP:V.

While the Druckman & Swets (1988) source does review NLP’s assumptions from a psychological perspective, it does not directly address its therapeutic effectiveness. It is not a systematic review, meta analysis or critical review of ‘’’its therapeutic’’’ effectiveness. So my suggestion was to use separate, relevant systematic review, critical review or meta analyses to substantiate each each part of the statement in line with WP:MEDRS. The textbook you mentioned (that had a section critiquing the use of NLP in influence) would not meet that criteria either but would also require page numbers for verifiability, and it is not a systematic review.

Statement #1 makes specific claims about NLP’s assumptions and its therapeutic effectiveness, which are distinct aspects of NLP. Therefore, it’s crucial to ensure that the sources cited for this statement directly support the respective claims in line with WP:NOR. —Notgain (talk) 00:21, 11 May 2024 (UTC)

So we agree but we don't agree... I still don't get it.

"It is not a systematic review, meta analysis or critical review of ‘’’its therapeutic’’’ effectiveness.", and how is that a problem?, did you even notice that is not the only source listed in the specific note (which is the k one) for those affirmations right?

As i said, it would be a problem if it was the only source for such affirmations. Which is not the case. Rodrigo IB (talk) 03:13, 11 May 2024 (UTC)

The use of endnote [k] to reference Statement #1 (S1) without page numbers makes it difficult to confirm if the claims are supported. Its unclear which source supports which part of the statement raising issue of WP:OR. The statement mentions "numerous literature reviews and meta-analyses," yet none of the six references in [k] are meta-analyses, so it is misleading. Witkowski (2010) is the only more recent quantitative and qualitative literature review of the empirical evidence (there are more recent ones that could be added). Sharpley (1983/87) and Heap (1988) focuses on the contested PRS. Heap (1988) explicitly states that NLP's effectiveness in clinical settings had yet to be experimentally evaluated at that time. Von Bergen et al. (1997) is unrelated, focusing on NLP in human resources development (HRD) - there are more recent review related to HRD. So I suggest page numbers should be added, and the relevance of each source to the statement should be clarified. Modifying Statement #1 to accurately reflect the cited sources and potentially incorporating additional, relevant meta-analyses or systematic reviews. --Notgain (talk) 12:40, 11 May 2024 (UTC)

Witkowski 2010 is a meta-analysis.

The sources themselves cite other studies and meta-analysis which aren't as accessible (the use of public access sources is something that we as editors must try to implement, there are instances in which a reliable source is behind a paywall and shouldn't be discarted. This aspect kind of limits the sources that can be used by Wikipedia, more on that here: WP:Reliable sources/Cost.) Rodrigo IB (talk) 16:51, 11 May 2024 (UTC)

Witkowski 2010 is a literature review, not a meta-analysis. The paywall issue, while important, doesn’t address the relevance and accuracy of sources. Druckman & Swets (1988) doesn’t directly address therapeutic effectiveness, so it may not be the most suitable reference for that part of the statement. —Notgain (talk) 20:38, 11 May 2024 (UTC)

The wikipedia article of Meta-analysis as well: "Meta-analyses are often, but not always, important components of a systematic review procedure."

The wikipedia article for systematic reviews: "In practice, when one is mentioned the other may often be involved, as it takes a systematic review to assemble the information that a meta-analysis analyzes, and people sometimes refer to an instance as a systematic review even if it includes the meta-analytical component. " Which is the case for Witkowski 2010 (which is not a literature review as you said), and is presented in the page 60.

I didn't explained myself well on the subject of accessibility, sorry for that; but i brought it to the table because we are also discussing citation problems within the article and the changes you have been trying to do. The thing is that we cannot put more and more sources for a series of affirmations. For that we need to follow certain guidelines like the mentioned WP:Reliable sources/Cost, to ensure the WP:V, and WP:NOR, the other issue is that the changes you did would infringe not just the previous citation style but the "orientation" sort to speak of editors of what sources are public, hard to verify (like offline sources) or behind a paywall. This is important because it could help improve the article if an affirmation hasn't been verified.

The sources more than just once conclude with the fact that NLP lacks empirical evidence. There is no original research problem in such affirmations. Rodrigo IB (talk) 21:39, 11 May 2024 (UTC)

With all due respect, Witkowski 2010 is a critical analysis, not a systematic review. It lacks a pre-defined protocol, specific research question, and rigorous assessment of evidence quality, which are key characteristics of a systematic review. See WP:MEDRS—01:03, 12 May 2024 (UTC) Notgain (talk) 01:03, 12 May 2024 (UTC)

Witkowski 2010 —Notgain (talk) 01:07, 12 May 2024 (UTC)

If the statement #1 is claiming that “numerous literature reviews and meta-analyses have failed to show evidence for NLP’s assumptions or effectiveness as a therapeutic method,” then it should be supported by references to actual literature reviews and meta-analyses. There are no meta analyses directly cited, so either add a citation if it exists and meets WP:RS, or revise the statement for accuracy. —Notgain (talk) 01:28, 12 May 2024 (UTC)

The systematic review of Witkowski is rigorous enough in its analysis of the quality of the presented evidence. If you have any concern in such aspect then clarify it, be specific for those concerns.

There is no affirmation that violates WP:NOR with the cited sources. Rodrigo IB (talk) 02:40, 12 May 2024 (UTC)

In the letter to the editor of the Polish Psychological Bulletin, Aleksandra cite Witkowski 2010 as a "systematic review" Witkowski, Tomasz; Luszczynska, Aleksandra (2013-12-01). "Letters to Editor". Polish Psychological Bulletin. 44 (4): 462–464. doi:10.2478/ppb-2013-0049. ISSN 0079-2993. along with Sturt 2012 doi:10.3399/bjgp12X658287. So at least one third party source refer to it as a systematic review. However, it does not meet the PRISMA criteria for a systemic review and there is no statistical meta-analysis. So I still maintain it would be better described a critical review of empirical research. There are a number of systematic reviews that came after Witkowski 2010 as we discussed earlier. And I think there is at least one 2015 meta-analysis, Zaharial, Reiner and Schütz 2015 PMID 26609647 that has not been cited yet. --Notgain (talk) 11:24, 12 May 2024 (UTC)

The meta-analysis of Schütz et al is flawed. I'm reading it [4] and is worrying that the total of studies that were analyzed were just 12.

But there is another issue.

"Overall, we finally included 12 studies with a total number of individuals of 658 (studies that analysed different subgroups from the same population). On average, the numbers of participants in each study was small, ranging between 12 and 115 subjects".

One component of the inclusion-exclusion criteria: "Not the right population: studies conducted on healthy individuals with social/psychological problems (n=19)"

Data analysis: "The inspection of the funnel plot was done visually."

Jeffrey Chan and Amer Harky [5] warn of the inclusion of non-randomized studies without risk of bias assessment (I mention it too because the small commentary also mentions the risks involved in methods that use visual inspection of heterogeneity across studies).

Schwarzer et al. [6] give a more comprehensive picture of the risks sorrounding meta-analysis that use small studies. Like the one you cite. Im well aware that Schütz et al. conducted a publication bias analysis: "Begg and Majumdar's rank correlation nor Egger’s regression test was significant (p=0.73 and p=0.45, respectively), which indicates no publication bias."

But, as Schwarzer et al. Point out other possible causes: "Another possible cause of small-study effects is clinical heterogeneity between patients in large and small studies; e.g., patients in smaller studies may have been selected so that a favourable outcome of the experimental treatment may be expected. In the case of a binary outcome, also a mathematical artefact arises from the fact that for the odds ratio or the risk ratio, the variance of the treatment effect estimate is not independent of the estimate itself [...] Lastly, it can never be ruled out that small-study effects result from mere coincidence [42]. Empirical studies have established evidence for these and other kinds of bias [19, 42, 53]. There is a vast range of tests for small-study effects [4, 20, 24, 38, 43, 48], most of them based on a funnel plot which will be introduced in Sect. 5.1.1"

My concerns is that the meta-analysis you brought to the table is a false-positive, even the authors write: "there is a major lack of high-quality data from observational, experimental studies or randomized trials on this field, Up until now there is insufficient data to recommend this form of therapy strongly in reducing some psychosocial problems." Making it an inconclusive study. Rodrigo IB (talk) 20:04, 12 May 2024 (UTC)

Let’s not cherry pick. I return to the Statement #1 that currently mentions “numerous literature reviews and meta-analyses”. The limitations of these studies should be mentioned if in line with WP:NOR —Notgain (talk) 23:57, 12 May 2024 (UTC)

Cherry pick what?

Are you really trying to reach a consensus or not? because i don't see you actually addressing the points that emerge of the sources that you, you as a proponent of them should be considering and analyzing in a careful manner.

You know what's the worst?, that i shouldn't have made that analysis, not only because is your responsability to at least read the sources you want to implement, but because the page talk is not for that. Neither of what is your opinion of what is or what is not a systematic review. As you said, reliable sources refer to Witkowski 2010 as a systematic review. End of the debate.

There is no original research involved, period. Rodrigo IB (talk) 02:34, 13 May 2024 (UTC)

That was an astute observation. The funnel plot inspection in the Schütz et al. study was done visually, which can introduce subjectivity and potential bias. While they noted this limitation and the small number of studies, this impacts the robustness of findings. This is amplified because the authors (e.g. Peter Schütz) appear to be practitioners (not academic researchers) which introduces another source of potential subjectivity bias. If it were to be cited, the limitations would need to be made clear. I maintain that Witkowski 2010 is not a systematic review or meta-analysis - it was a scathing critical review of empirical literature. It does NOT meet the PRISMA criteria for systematic review or meta-analysis as noted earlier. I encourage you to consider these points and reevaluate. If you have evidence to the contrary, I would be interested to hear it. --Notgain (talk) 09:28, 13 May 2024 (UTC)

Why you keep insisting on the PRISMA declaration?.

I don't know if you are aware, but the PRISMA declaration was sort of an "update" to the QUORUM declaration in 2009. Which it's main focus at the beginning was clinical meta-analysis and systematic reviews. It was not as adopted in 2010 like now, even the paper presenting the declaration was published at the beginning of 2010. [7]

Still tho, Witkowski meets the QUORUM declaration. But the declaration is not necessary in order to consider something as a meta-analysis or a systematic review. It just secures that the data analysis is not biased in certain ways. Rodrigo IB (talk) 15:03, 13 May 2024 (UTC)

Morgan 1993

Removed the following from further reading because it is impossible to find and its outdated or near impossible to find: Morgan, Dylan A. (1993). "Scientific Assessment of NLP". Journal of the National Council for Psychotherapy & Hypnotherapy Register. Spring. 1993. --Notgain (talk) 09:42, 6 May 2024 (UTC)

This has been discussed in the past in Talk:Neuro-linguistic programming/Archive_4#Morgan and Heap. The consensus was to replace any citations with citations to Heap. But that was already done and Morgan isn't used, so I agree removing it from Further reading is fine. Skyerise (talk) 11:15, 6 May 2024 (UTC)

Looks like it... These were the Heap papers: [8]

[9][10][11][12] --Notgain (talk) 13:16, 6 May 2024 (UTC)