Talk:Glyphosate/Archive 17
 | This is an archive of past discussions about Glyphosate. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
| Archive 10 | ← | Archive 15 | Archive 16 | Archive 17 | Archive 18 | Archive 19 | Archive 20 |
Acceptable reviews
After leaving all Monsanto articles some time ago I took hope that things have changed here. I was wrong and hope that I do not have to again leave. Using this review [1] because it had the (Redacted) my added information was still deleted saying that it did not meet our WP:MEDRS standards. It is very discouraging to spend that much time reading a review and adding what I believed to be appropriate information only to have it deleted. Gandydancer (talk) 22:16, 10 May 2019 (UTC)
- I used that source specifically because KingofAces said it was WP:MEDRS. How that is called an attack that needed to be redacted is beyond me. Gandydancer (talk) 23:54, 10 May 2019 (UTC)
- I can't comment on what KingofAces said but there are a few things to bear in mind: As I'm sure has been discussed here before, articles in Frontiers journals can be questionable and we should be careful how we use them. This specific one is a bit strange in that the article reviewers are all authors who are cited extensively in the review and have some relatively fringe views in this area (similar to Seralini). Other researchers have commented that the fatty liver disease work contained "a number of questionable issues related to the experimental design and the interpretation of the obtained data." The idea of MEDRS is to get around problems of interpreting the findings of individual pieces of primary research. If as in this case though, the reviewers of the review are themselves the authors of the primary research, it's fairly obvious that they are not going to critically evaluate their findings. There's a considerable body of work on the NHL risk, but the fatty liver disease work comes from a single paper and so shouldn't be presented as if ti is equivalent. SmartSE (talk) 14:07, 11 May 2019 (UTC)
- Thanks for your work on this issue. I did not choose that article but rather used it because KoA had said that it was MEDRS compliant (which actually surprised me). All of your criticisms seem reasonable. Gandydancer (talk) 14:58, 11 May 2019 (UTC)
- I can't comment on what KingofAces said but there are a few things to bear in mind: As I'm sure has been discussed here before, articles in Frontiers journals can be questionable and we should be careful how we use them. This specific one is a bit strange in that the article reviewers are all authors who are cited extensively in the review and have some relatively fringe views in this area (similar to Seralini). Other researchers have commented that the fatty liver disease work contained "a number of questionable issues related to the experimental design and the interpretation of the obtained data." The idea of MEDRS is to get around problems of interpreting the findings of individual pieces of primary research. If as in this case though, the reviewers of the review are themselves the authors of the primary research, it's fairly obvious that they are not going to critically evaluate their findings. There's a considerable body of work on the NHL risk, but the fatty liver disease work comes from a single paper and so shouldn't be presented as if ti is equivalent. SmartSE (talk) 14:07, 11 May 2019 (UTC)
Is there any objection to using either of these sources? [2] and [3] ? Gandydancer (talk) 17:21, 11 May 2019 (UTC)
- I think the first of those two is an excellent find, thanks! We should definitely use it. The second one is more tricky, because it's labeled by the journal as an "essay" (see here), so it isn't really a review, so much as some recommendations. It's probably best suited for sections that discuss regulation, rather than those about the toxicity ("safety standards" as opposed to "safety"). --Tryptofish (talk) 17:33, 11 May 2019 (UTC)
- The second one is a review as well.
- See: "Publication types Review Research Support, Non-U.S. Gov't Gandydancer (talk) 18:05, 11 May 2019 (UTC)
- Yes, I saw that, and I take it as meaning a review of safety standards, as opposed to a review of the scientific literature. Here: [4], is a link to how the paper actually appeared in the journal, and it's very clearly labeled "Essay". I think it's fine as a source in sections about regulation, to say something approximately like "More recently, Vandenberg et al. have recommended changing the safety standards by..." But I would not use it for saying that glyphosate isn't as safe as people used to think. That's the distinction I'm making. --Tryptofish (talk) 18:33, 11 May 2019 (UTC)
- My edit read: Noting a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and that current safety assessments rely heavily on studies done over 30 years ago, a 2018 review found current safety standards to be outdated and "may fail to protect public health or the environment."[85] According to our article there is in fact a 100% increase in use and mixtures are probably more harmful than glyphosate alone. I don't know about the "30 years" part but it sounds about right. I do not believe that to say in our article that the authors say current standards "may fail to protect public health or the environment." would violate our guidelines. And again, this was published on PubMed as a review article. This was rejected by Kingofaces due to Best not to use a Benbrook et al. paper as discussed previously, and not a review in terms of WP:MEDRS (essays are often of much lower quality than even primary research articles) Also note that you and another editor now apparently have no problem using Benbrook. There seems to be a double standard problem here. Gandydancer (talk) 20:36, 11 May 2019 (UTC)
- I agree with Tryptofish in that it is an essay piece rather than a classical review and that it is more relevant to regulation and future research than current opinions on safety. The hundred-fold part seems unrelated to the issue of safety and in fact this part of the lead also seems a bit over the top given that it was only introduced in 1974 (it's a million times more than at some point!). It would be more relevant to frame it in terms of when the roundup ready crops were introduced since this was the main driver of increased use (although being off-patent and the move towards no-till farming are also drivers). Bit tangential to this mind! As for Benbrook, as I thought I made clear earlier, it depends on the source and how the source is used. Unfortunately things are not black and white. I don't think there will be any objections to adding content saying "more research is required" but to make it seem as if everyone agrees that the current safety standards are outdated is not appropriate and framing it as a review does give this impression, whereas saying it is essay (i.e. opinion) would make that clearer. SmartSE (talk) 21:15, 11 May 2019 (UTC)
- Smartse articulates very well what I was trying to say about how to use and how not to use that source. I would be OK with something approximately like: "Vandenberg et al. have noted that current safety assessments rely heavily on studies that are over 30 years old, and recommend that the standards be made more strict." I'll leave it to Kingofaces to explain his thinking. --Tryptofish (talk) 21:29, 11 May 2019 (UTC)
- Again, PubMed calls it a review. Per MEDRS "Broadly speaking, reviews may be narrative or systematic (and sometimes both). Narrative reviews often set out to provide a general summary of a topic based on a survey of the literature, which can be useful when outlining a topic." IMO this is clearly a narrative review. Re "The hundred-fold part seems unrelated to the issue of safety", I can't understand why the amount in the environment would not be related to the issue of safety. Several articles that I have read have made similar comments about both the increase of use, combinations of products used due to weed resistance, and the new method of using the product to finish off a field in a more uniform pattern, leaving higher levels on the product. Gandydancer (talk) 22:51, 11 May 2019 (UTC)
- Although I suggested some wording above, I actually don't feel strongly either way about how we source the amount in the environment. Maybe other editors feel more strongly than I do. --Tryptofish (talk) 00:28, 12 May 2019 (UTC)
- SmartSE said, "I don't think there will be any objections to adding content saying "more research is required" but to make it seem as if everyone agrees that the current safety standards are outdated is not appropriate and framing it as a review does give this impression, whereas saying it is essay (i.e. opinion) would make that clearer." We will need to disagree on whether or not it is a narrative review but I do feel that we need to include their conclusion/opinion: "We conclude that current safety standards for GBHs are outdated and may fail to protect public health or the environment." It can be called an essay if that satisfies your concerns. Gandydancer (talk) 11:39, 13 May 2019 (UTC)
- Although I suggested some wording above, I actually don't feel strongly either way about how we source the amount in the environment. Maybe other editors feel more strongly than I do. --Tryptofish (talk) 00:28, 12 May 2019 (UTC)
- Again, PubMed calls it a review. Per MEDRS "Broadly speaking, reviews may be narrative or systematic (and sometimes both). Narrative reviews often set out to provide a general summary of a topic based on a survey of the literature, which can be useful when outlining a topic." IMO this is clearly a narrative review. Re "The hundred-fold part seems unrelated to the issue of safety", I can't understand why the amount in the environment would not be related to the issue of safety. Several articles that I have read have made similar comments about both the increase of use, combinations of products used due to weed resistance, and the new method of using the product to finish off a field in a more uniform pattern, leaving higher levels on the product. Gandydancer (talk) 22:51, 11 May 2019 (UTC)
- Smartse articulates very well what I was trying to say about how to use and how not to use that source. I would be OK with something approximately like: "Vandenberg et al. have noted that current safety assessments rely heavily on studies that are over 30 years old, and recommend that the standards be made more strict." I'll leave it to Kingofaces to explain his thinking. --Tryptofish (talk) 21:29, 11 May 2019 (UTC)
- I agree with Tryptofish in that it is an essay piece rather than a classical review and that it is more relevant to regulation and future research than current opinions on safety. The hundred-fold part seems unrelated to the issue of safety and in fact this part of the lead also seems a bit over the top given that it was only introduced in 1974 (it's a million times more than at some point!). It would be more relevant to frame it in terms of when the roundup ready crops were introduced since this was the main driver of increased use (although being off-patent and the move towards no-till farming are also drivers). Bit tangential to this mind! As for Benbrook, as I thought I made clear earlier, it depends on the source and how the source is used. Unfortunately things are not black and white. I don't think there will be any objections to adding content saying "more research is required" but to make it seem as if everyone agrees that the current safety standards are outdated is not appropriate and framing it as a review does give this impression, whereas saying it is essay (i.e. opinion) would make that clearer. SmartSE (talk) 21:15, 11 May 2019 (UTC)
- My edit read: Noting a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and that current safety assessments rely heavily on studies done over 30 years ago, a 2018 review found current safety standards to be outdated and "may fail to protect public health or the environment."[85] According to our article there is in fact a 100% increase in use and mixtures are probably more harmful than glyphosate alone. I don't know about the "30 years" part but it sounds about right. I do not believe that to say in our article that the authors say current standards "may fail to protect public health or the environment." would violate our guidelines. And again, this was published on PubMed as a review article. This was rejected by Kingofaces due to Best not to use a Benbrook et al. paper as discussed previously, and not a review in terms of WP:MEDRS (essays are often of much lower quality than even primary research articles) Also note that you and another editor now apparently have no problem using Benbrook. There seems to be a double standard problem here. Gandydancer (talk) 20:36, 11 May 2019 (UTC)
- Yes, I saw that, and I take it as meaning a review of safety standards, as opposed to a review of the scientific literature. Here: [4], is a link to how the paper actually appeared in the journal, and it's very clearly labeled "Essay". I think it's fine as a source in sections about regulation, to say something approximately like "More recently, Vandenberg et al. have recommended changing the safety standards by..." But I would not use it for saying that glyphosate isn't as safe as people used to think. That's the distinction I'm making. --Tryptofish (talk) 18:33, 11 May 2019 (UTC)
- See: "Publication types Review Research Support, Non-U.S. Gov't Gandydancer (talk) 18:05, 11 May 2019 (UTC)
- The second one is a review as well.
I think the best thing at this point is to get some outside editor opinions, so I have started WP:RSN#Distinguishing a review from an essay. --Tryptofish (talk) 16:32, 13 May 2019 (UTC)
- Gandydancer, when did I say the source was reliable? I've more or less mirrored what SmartSE said above about Frontiers journals, and that hasn't changed. It's possible I missed that this one was such a journal, but I don't recall where that would have been. Kingofaces43 (talk) 14:47, 14 May 2019 (UTC)
- After you removed my article info saying the source was of poor quality I used a source you mentioned related to Monsanto study funding in which they found "no criticism of methodology, findings, etc." so as to be certain not waste my time all over again. Perhaps ironic that that was the one that turned out to be of possible low quality and not the one I had used. Gandydancer (talk) 14:49, 15 May 2019 (UTC)
- So basically, you claimed I said something I did not say at all. In that case, we were only discussing if there had been any criticism in any studies citing the study (which was no), not that those citing sources were appropriate for use here. Those are two very different discussions. Kingofaces43 (talk) 00:13, 16 May 2019 (UTC)
- I did nothing of the sort. When you desired to show that there had been no complaints re Monsanto funding you found that study to be both MEDRS and capable of making a Monsanto judgement. I don't know how many times I need to say this - I did not set out to prove you wrong, cause debate, or any thing other than to attempt to use a source that would find approval here, something that has proven to not be an easy thing to do. Gandydancer (talk) 01:07, 16 May 2019 (UTC)
you found that study to be both MEDRS
. That is exactly the sort, so please stop insisting I said such when I reiterated multiple times before this that it was not MEDRS. That conversation only got to the point of saying there were two reviews that cited the study in question and they didn't say anything on funding. That particular avenue stopped short of any reliability discussion because there was no relevant content to discuss in the first place. Put another way, even low-quality sources didn't have anything to say on the matter much less higher quality ones. Kingofaces43 (talk) 14:08, 17 May 2019 (UTC)
- I did nothing of the sort. When you desired to show that there had been no complaints re Monsanto funding you found that study to be both MEDRS and capable of making a Monsanto judgement. I don't know how many times I need to say this - I did not set out to prove you wrong, cause debate, or any thing other than to attempt to use a source that would find approval here, something that has proven to not be an easy thing to do. Gandydancer (talk) 01:07, 16 May 2019 (UTC)
- So basically, you claimed I said something I did not say at all. In that case, we were only discussing if there had been any criticism in any studies citing the study (which was no), not that those citing sources were appropriate for use here. Those are two very different discussions. Kingofaces43 (talk) 00:13, 16 May 2019 (UTC)
- I think it's useful to look critically at the differences between narrative and systematic reviews. I'll reproduce here an image that is at WP:MEDRS. It shows that the two kinds of reviews serve very different levels of evaluating scientific evidence. In the hierarchy of source reliability, narrative reviews aren't that different from editorials. From WP:MEDRS#Biomedical journals:
Narrative reviews often set out to provide a general summary of a topic based on a survey of the literature, which can be useful when outlining a topic. Systematic reviews use sophisticated methodology to address a particular clinical question in as balanced (unbiased) a way as possible.
When the issues are controversial, the degree of being unbiased becomes particularly important for our purposes here. --Tryptofish (talk) 17:12, 15 May 2019 (UTC)
- This is misleading - you've cut CFCF's visual in half, left out his wording and added your own. With the visual you have added it appears that narrative reviews are very low quality sources. Actually the article does not say that nor do the two editors at WP:RSN#Distinguishing a review from an essay. In fact those two editors are part of the group that wrote the guidelines many years ago and could be considered capable of giving expert opinions. You should remove this misleading visual that only appears to validate your opinion rather than being informative. Gandydancer (talk) 13:10, 17 May 2019 (UTC)
- That visual comes directly from MEDRS, so there's nothing misleading, especially if you look at the context there. Narrative reviews generally are near bottom of the barrel in terms of secondary sources, and sometimes can be even more prone to issues than primary studies when there is no design as the visual points out. Within narrative reviews, you can have authors do an exhaustive literature search that gets close to being a systematic review to those cherry-picking only the ideas they want to present. In short, there can be a ton of variation in narrative reviews, which is why cautions get brought up about them compared to systematic reviews. Kingofaces43 (talk) 14:08, 17 May 2019 (UTC)
- The visual that Tryptofish has copied here shows us how Proctor and Gamble classifies information, not how Wikipedia classifies it. It is used as an example of the different ways to rank the importance of information. Gandydancer (talk) 13:31, 18 May 2019 (UTC)
- Gandydancer, the reason that I only reproduced one half of that image is that the other half does not include "narrative reviews", so I don't think that we could get any information about narrative reviews from that. And everything that I quoted from MEDRS is a direct quote. Uninvolved editors at RSN have pretty much rejected the idea that the source is a scientific review, with the range of opinions there extending from saying that it is purely an opinion piece, to saying that it is somewhat more of an essay than a review, with properties of both. You are the only editor arguing that it is purely a narrative review and not an essay. --Tryptofish (talk) 17:10, 17 May 2019 (UTC)
- Please review the conversation at the RSN page and note that I did not argue that it is a review but rather noted that PubMed termed it a review. I was perfectly satisfied to note that the article is the opinion of the authors. Gandydancer (talk) 13:31, 18 May 2019 (UTC)
- That visual comes directly from MEDRS, so there's nothing misleading, especially if you look at the context there. Narrative reviews generally are near bottom of the barrel in terms of secondary sources, and sometimes can be even more prone to issues than primary studies when there is no design as the visual points out. Within narrative reviews, you can have authors do an exhaustive literature search that gets close to being a systematic review to those cherry-picking only the ideas they want to present. In short, there can be a ton of variation in narrative reviews, which is why cautions get brought up about them compared to systematic reviews. Kingofaces43 (talk) 14:08, 17 May 2019 (UTC)
- This is misleading - you've cut CFCF's visual in half, left out his wording and added your own. With the visual you have added it appears that narrative reviews are very low quality sources. Actually the article does not say that nor do the two editors at WP:RSN#Distinguishing a review from an essay. In fact those two editors are part of the group that wrote the guidelines many years ago and could be considered capable of giving expert opinions. You should remove this misleading visual that only appears to validate your opinion rather than being informative. Gandydancer (talk) 13:10, 17 May 2019 (UTC)
Comparing some wording choices
At WP:RSN#Distinguishing a review from an essay, editors have proposed various ways to word what we might put on the page about the Vandenberg source. I'd like to outline some of the discussion so far, here, and ask about the advantages and disadvantages of various options.
Earlier, Gandydancer suggested this wording:
- Noting a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and that current safety assessments rely heavily on studies done over 30 years ago, a
20182019 review found current safety standards to be outdated and "may fail to protect public health or the environment."
- Noting a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and that current safety assessments rely heavily on studies done over 30 years ago, a
(I've corrected the date from 2018 to 2019.)
More recently, Gandydancer suggested this revision:
- Noting a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and that current safety assessments rely heavily on studies done over 30 years ago, in 2019 Vandenberg et al concluded that current safety standards are outdated and "may fail to protect public health or the environment."
I'd like to suggest, instead, this revision, in which I try to pay close attention to WP:CLAIM:
- In 2019 Vandenberg et al. cited a 100-fold increase in the use of glyphosate-based herbicides from 1974 to 2014, the possibility that herbicide mixtures likely have effects that are not predicted by studying glyphosate alone, and reliance of current safety assessments on studies done over 30 years ago. They recommended that current safety standards be updated, writing that the current standards "may fail to protect public health or the environment."
--Tryptofish (talk) 17:27, 17 May 2019 (UTC)
- I think that's preferable, both because it's an active sentence and because it avoids WP:CLAIM issues. Alexbrn (talk) 17:32, 17 May 2019 (UTC)
- I am more than satisfied with Trypto's version. It is more well-written and sounds more "encyclopedic". Gandydancer (talk) 13:34, 18 May 2019 (UTC)
- Thanks very much! I am happy that we were able to find a version that we can agree on. --Tryptofish (talk) 17:48, 18 May 2019 (UTC)
- I am more than satisfied with Trypto's version. It is more well-written and sounds more "encyclopedic". Gandydancer (talk) 13:34, 18 May 2019 (UTC)
- I think that's preferable, both because it's an active sentence and because it avoids WP:CLAIM issues. Alexbrn (talk) 17:32, 17 May 2019 (UTC)
- Late to the party as I haven't been around much... As I noted earlier I think the 100-fold increase part is irrelevant to this, mainly because it was only introuced in 1974. I don't dispute that the source mentions this, but it doesn't seem the most pertinent point which is rather that the studies were done a long time ago and many were not peer reviewed. We should seek to narrate how and why use of glyphosate has increased over time in Glyphosate#Uses but it seems superfluous in this section. Thoughts? SmartSE (talk) 17:22, 29 May 2019 (UTC)
- I'll go along with anything that gets consensus about it, but I feel like it was exhausting to get where we are now, and I wouldn't want to reopen that without a strong consensus to do so. --Tryptofish (talk) 21:20, 29 May 2019 (UTC)
- Understandable and I'm not going to argue incessantly about it. SmartSE (talk) 21:48, 29 May 2019 (UTC)
- Thanks, no problem. --Tryptofish (talk) 22:17, 29 May 2019 (UTC)
- I'll agree with removing the 100-fold bit as it wasn't really the focus on the above discussion. I was going to leave it be because of the exhaustion aspect you mentioned, but it should be a simple enough change without changing the core meaning.
- That being said, the Glyphosate#Uses section is where that concept could be addressed more someday. I saw some literature about how glyphosate replaced older more toxic herbicides over time that could fit well in that section, but I'd have to go digging for those again since I don't save citations in this subject like I used to. That nuance is usually needed when discussing changes in herbicide use, so it would probably be better to compartmentalize that discussion there instead. Kingofaces43 (talk) 22:29, 29 May 2019 (UTC)
- Understandable and I'm not going to argue incessantly about it. SmartSE (talk) 21:48, 29 May 2019 (UTC)
- I'll go along with anything that gets consensus about it, but I feel like it was exhausting to get where we are now, and I wouldn't want to reopen that without a strong consensus to do so. --Tryptofish (talk) 21:20, 29 May 2019 (UTC)
- Late to the party as I haven't been around much... As I noted earlier I think the 100-fold increase part is irrelevant to this, mainly because it was only introuced in 1974. I don't dispute that the source mentions this, but it doesn't seem the most pertinent point which is rather that the studies were done a long time ago and many were not peer reviewed. We should seek to narrate how and why use of glyphosate has increased over time in Glyphosate#Uses but it seems superfluous in this section. Thoughts? SmartSE (talk) 17:22, 29 May 2019 (UTC)
New meta-analysis out today says "raises risk of non-Hodgkin lymphoma by 41%"
https://www.sciencedirect.com/science/article/pii/S1383574218300887
"Using the highest exposure groups when available in each study, we report the overall meta-relative risk (meta-RR) of NHL in GBH-exposed individuals was increased by 41% (meta-RR = 1.41, 95% CI, confidence interval: 1.13–1.75)." TimidGuy (talk) 15:23, 14 February 2019 (UTC)
- There's already text stating the scientific consensus is that glyphosate does not pose a significant carcinogenic risk at Glyphosate#Cancer. Pretty much every regulatory agency, etc. agrees, and stronger meta-analyses than this new study show no risk too. Other studies on NHL and glyphosate links have been criticized for methodological issues, cherry-picking, confounding, etc. in reviews, so I don't think there's anything that could really be changed in the article based on this new study in terms of WP:DUE. It could be worth fleshing out what other reviews have said with respect to issues in the studies looking at NHL though since it seems to come up often. Kingofaces43 (talk) 17:30, 14 February 2019 (UTC)
- The deletion of this new research seems strange. If the article passed academic scrutiny, it by definition must provide new evidence that could change scientific opinion. The fact that there swas "scientific consensus" prior to this research does not mean the consensus cannot change once a new research is published. How can a Wikipedia editor be more knowledgeable than the academic reviewers of this paper? — Preceding unsigned comment added by 84.109.189.248 (talk) 21:33, 15 February 2019 (UTC)
- I reverted the edit by 84.109, because it was sourced to an announcement from the university website, which is tantamount to a press release. I do realize that, just above, TimidGuy links to the actual journal article, and that the article is a meta-analysis. This has nothing to do with what Wikipedia editors know (and we have a policy against basing such decisions on what editors think that they know). Rather, because we are discussing a disease in human beings, the decisions about sourcing must be based on WP:MEDRS. And that rules out reporting something because it is "breaking news" in research. As Kingofaces correctly points out, this is one meta-analysis among multiple meta-analyses and review articles, and there is a high barrier to using it to effectively overrule the other sources. We don't know yet whether or not scientific consensus has changed. (The fact that the new paper could change scientific consensus isn't good enough: guessing whether something could do that is just editor opinion.) If it does change consensus, that will soon be reflected in other publications that are independent of this one, and if/when that happens, Wikipedia most certainly will report it. --Tryptofish (talk) 00:06, 16 February 2019 (UTC)
- IMO this reflects a very strange situation in U.S. Science (I am not American). (1) I would expect a university press release (which includes a link to the research paper) to be identical to the research findings, except that its stated in a layman's terms. It seems that you view it as an advertisement, i.e. something whose credibility should be questioned and whose purpose is (at least in part) to hype the university's status. (2) I would expect an academic article to advance science, i.e. to provide further evidence than known before. In such a case, I would have thought that Wikipedia should reflect that new information exists. It seems that you see a new article as merely a statement of fact (hopefully you don't question its correctness), without any say about its relevance. I wonder - in what cases do you thing the scientific community would reject the results of such a paper: wrong methodology? not enough analyses gathered for a meta-analysis? Why didn't the paper's reviewers reject it on these grounds? (3) Who decides what is the "scientific consensus" - Is this an NIH/WHO decision? -- Ronnie — Preceding unsigned comment added by 84.109.189.248 (talk) 18:22, 16 February 2019 (UTC)
- This isn't about how US science works. It's about how Wikipedia works. If you want to change how Wikipedia works in this instance, the place to discuss it is at WT:MEDRS. --Tryptofish (talk) 20:19, 16 February 2019 (UTC)
- Thanks, I opened a discussion there. — Preceding unsigned comment added by 84.109.189.248 (talk) 20:52, 16 February 2019 (UTC)
- This isn't about how US science works. It's about how Wikipedia works. If you want to change how Wikipedia works in this instance, the place to discuss it is at WT:MEDRS. --Tryptofish (talk) 20:19, 16 February 2019 (UTC)
- IMO this reflects a very strange situation in U.S. Science (I am not American). (1) I would expect a university press release (which includes a link to the research paper) to be identical to the research findings, except that its stated in a layman's terms. It seems that you view it as an advertisement, i.e. something whose credibility should be questioned and whose purpose is (at least in part) to hype the university's status. (2) I would expect an academic article to advance science, i.e. to provide further evidence than known before. In such a case, I would have thought that Wikipedia should reflect that new information exists. It seems that you see a new article as merely a statement of fact (hopefully you don't question its correctness), without any say about its relevance. I wonder - in what cases do you thing the scientific community would reject the results of such a paper: wrong methodology? not enough analyses gathered for a meta-analysis? Why didn't the paper's reviewers reject it on these grounds? (3) Who decides what is the "scientific consensus" - Is this an NIH/WHO decision? -- Ronnie — Preceding unsigned comment added by 84.109.189.248 (talk) 18:22, 16 February 2019 (UTC)
- I reverted the edit by 84.109, because it was sourced to an announcement from the university website, which is tantamount to a press release. I do realize that, just above, TimidGuy links to the actual journal article, and that the article is a meta-analysis. This has nothing to do with what Wikipedia editors know (and we have a policy against basing such decisions on what editors think that they know). Rather, because we are discussing a disease in human beings, the decisions about sourcing must be based on WP:MEDRS. And that rules out reporting something because it is "breaking news" in research. As Kingofaces correctly points out, this is one meta-analysis among multiple meta-analyses and review articles, and there is a high barrier to using it to effectively overrule the other sources. We don't know yet whether or not scientific consensus has changed. (The fact that the new paper could change scientific consensus isn't good enough: guessing whether something could do that is just editor opinion.) If it does change consensus, that will soon be reflected in other publications that are independent of this one, and if/when that happens, Wikipedia most certainly will report it. --Tryptofish (talk) 00:06, 16 February 2019 (UTC)
- The deletion of this new research seems strange. If the article passed academic scrutiny, it by definition must provide new evidence that could change scientific opinion. The fact that there swas "scientific consensus" prior to this research does not mean the consensus cannot change once a new research is published. How can a Wikipedia editor be more knowledgeable than the academic reviewers of this paper? — Preceding unsigned comment added by 84.109.189.248 (talk) 21:33, 15 February 2019 (UTC)
I gotta say the framing of this discussion of how science works versus how WP works rubs me really wrong. Yes, WP is not a part of scientific process and the scientific process and debate has to happened outside WP and WP merely reflects the (settled) results or in (large) unresolved cases the current state of the debate. However Glyphosate is not Climatechanges and the above boldly claimed scientific consensus imho simply does not exist. And while it is true that many scientific organisations currently state that there is cancer risk associated with Glyphosate, it isn't true for all (see the IARC case). And while you can argue most studies don't show associations, it is again by far not true for all and imho if one were to declare a supposed consensus it is more along the line of "association with cancer is currently very weak, further research is needed", which is miles away from a supposed consensus a la "no association with cancer". The latter however seems to be used here to be able to argue that WP policy requires to dismiss this study (or any other new differing result) out of hand.--Kmhkmh (talk) 23:08, 16 February 2019 (UTC)
- "There's already text stating the scientific consensus..." A meta analysis has been posted. This isn't fringe content this is the most recent, MEDRS compliant information which adds to the growing scientific body of information on glyphosate and whatever its impact on health maybe. We overstep ourselves as editors when we decide that scientific consensus does not include the most recent meta analysis. We can decide weight and report this new meta anlysis carries less weight in terms of our articles, but we can't decide it doesn't exist in favor of an already determined POV. Again this isn't fringe content; we must report this, taking into account due weight, this new compliant content. I'm scratching my head on this.Littleolive oil (talk) 23:44, 16 February 2019 (UTC)
- I'd like to suggest a version of this."our current meta-analysis of human epidemiological studies suggests a compelling link between exposures to GBHs and increased risk for NHL.Littleolive oil (talk) 01:40, 17 February 2019 (UTC)
- That would go against the scientific consensus and violate WP:DUE (the opposite of "we must report this"). The scientific consensus is very clear that practically all national-level scientific groups and agencies agree with the consensus, and the lone outlier, the IARC, has been heavily criticized for its methodology. It's even to the point that the WHO, which the IARC is housed within, has ignored their designation and considers glyphosate not to be as significant carcinogen. Things going against the scientific consensus are indeed frequently WP:FRINGE (glyphosate and cancer claims frequently fall under that), but they are instead almost always undue weight until a consensus has actually changed. There's nothing even close to such as change yet, and a single meta-analysis cannot be used to contradict higher level sources. If those high-level scientific groups starting changing their designations, then there would be something to discuss. Until that happens though, Wikipedia cannot try to get ahead of the science.
- The other issue is that we've seen these kinds of meta-analyses on the subject already (see Glyphosate#Human. The problem is that the underlying literature is easily prone to bias and confounding, particularly on the non-Hodgkins lymphoma topic (the studies are correlational and frequently confounded with other pesticide use). There are weight issues with even including this study in the lower-tier of just looking at meta-analyses, so it's really become a topic where secondary coverage of the meta-analyses is frequently needed at this point. The status of that last paragraph I linked to stays the same in either case where some studies saw a correlation with NHL, but many find none with the positive associations often having methodological issues.
- Scientific consensus is already a high bar though. If it's climate change, we don't say we need to include sources that go against that consensus, we don't include studies saying there is a significant health risk to GMOs, etc. even if they are meta-analyses. Doing otherwise would contradict too many polices. A lot needs to happen for a study like this to be mentioned here. Kingofaces43 (talk) 05:22, 17 February 2019 (UTC)
- Again this is NOT chlimate change and the scientific consensus you claim simply does not exist.--Kmhkmh (talk) 05:39, 17 February 2019 (UTC)
- Please avoid violating WP:OR by claiming there is no consensus. The relevant portion of current text is very clear on this. A scientific consensus can occur in areas other than climate change. Kingofaces43 (talk) 06:23, 17 February 2019 (UTC)
- Again you claiming a scientific consensus, doesn't make one. And yes a scientific consensus is not restricted to climate change, in fact most of our science articles live in a consensus space. Glyphosate and various related issues however fail a scientific cosnsensus on many topics as far as I can see, instead you can only talk of majority opinions (in a heavily lobbied field). Also you can't really violate WP:OR on a talk page.--Kmhkmh (talk) 20:37, 17 February 2019 (UTC)
- Please avoid violating WP:OR by claiming there is no consensus. The relevant portion of current text is very clear on this. A scientific consensus can occur in areas other than climate change. Kingofaces43 (talk) 06:23, 17 February 2019 (UTC)
- Again this is NOT chlimate change and the scientific consensus you claim simply does not exist.--Kmhkmh (talk) 05:39, 17 February 2019 (UTC)
- Scientific consensus is already a high bar though. If it's climate change, we don't say we need to include sources that go against that consensus, we don't include studies saying there is a significant health risk to GMOs, etc. even if they are meta-analyses. Doing otherwise would contradict too many polices. A lot needs to happen for a study like this to be mentioned here. Kingofaces43 (talk) 05:22, 17 February 2019 (UTC)
- I've read through the paper (being summoned from
FTNWT:MEDRS) and I do think a sentence should go in the article at some point, perhaps after waiting a few weeks to make sure no major flaws are uncovered. We're already citing other NHL meta-analyses, and if there isn't anything wrong with this one then consistency would suggest adding it as well. The sources do seem pretty clear that there's a consensus with respect to general carcinogenicity, but my impression (I can't say for sure) is that this might indicate a minority view for NHL in particular that's significant enough to mention, again with the caveat that the paper's methods and results are not substantially questioned in the future. If not, then the presence of the other NHL meta-analyses should be re-evaluated.
- That said, if we want to consider additions, the text that was originally inserted and Olive's above proposal both contain editorializing as well as at least one major problem. For the latter version, the language "our current meta-analysis" suggests the most recent meta-analysis is necessarily the most authoritative, or even subsumes previously published meta-analyses, which are definitely not the case. For the former, it emphasizes the observed risk estimate when the authors specifically de-emphasize it and instead offer cautions on interpretation, and the use of the vague plural "some cancers" (while it could arguably be considered technically correct) gives the false impression of greater severity. (And on that note, while it's not explicitly laid out in this source, for context it may also be necessary to make the point that this is essentially focused on occupational exposures, which is very different from the environmental exposures that tend to be of greater interest to the general public.) Sunrise (talk) 05:54, 17 February 2019 (UTC)
- I’ve mentioned weight issues, and part of what you mention for other NHL text if somewhat historical. That is included because those initial NHL analyses caused a splash at the time, but were largely later discredited, hence the current layout. This source hasn’t reached that bar yet. Including itwould become more like dueling sources when the current version provides how other sources summarize the progression of literature instead. That background is a little buried within the sources if only glancing at text.
- As for where the consensus applies with respect to “general cancer”, the NHL studies were prominently weighed in on within the various consensus statements in the sources. They were looked at, and the weight of evidence was considered as no significant risk. That’s why we need to be careful about contradicting the consensus that includes NHL with a single meta-analysis when there has already been a lot of weighing in with previous sources. Kingofaces43 (talk) 06:22, 17 February 2019 (UTC)
- I couldn't find any mention of this at WP:FTN. Was it actually my post at WT:MEDRS? If not, Sunrise, could you please link to the discussion? Also, I believe that a thoughtful reading of WP:MEDRS does not lead to the overly simplistic conclusion that this is a MEDRS-compliant source, at least not if we present it in a way that gives it more weight than the source material as a whole. That doesn't mean that we are going to dismiss it out of hand. (Also, although no one is doing anything wrong here, I just want to remind everyone that the discretionary sanctions from the ArbCom GMO case apply to this page. Just giving the reminder before there is any problem.) --Tryptofish (talk) 19:48, 17 February 2019 (UTC)
- (For my part, I probably wouldn't have considered commenting here at all if I didn't know the atmosphere was being controlled by the DS! And yes, I meant WT:MEDRS, sorry for the mixup.) Your proposal below is in line with the sort of thing I was thinking of for the ideal case - minority (or fringe) positions, if mentioned, must be placed in the context of the majority view, and any addition would have to comply with that. However, I agree with the comment that this isn't coming across very well for the existing text. In particular, to me the weighting in the current version of the lead seems to imply there is at least a serious possibility of cancer risk: emphasis on "concerns" with no direct counterbalance, a lot of use of attribution and quotation, the 2014 meta-analysis getting the final word in paragraph 4, and IARC being given the most visible position in paragraph 5. The situation becomes clearer after reading the body, but the same kind of thing applies there too, such as giving the "limited evidence" statement the first position in the paragraph that describes the consensus statement. Sunrise (talk) 04:08, 18 February 2019 (UTC)
- Thanks Sunrise. Sorry if anything caused you any stress. I agree with you overall on the content issues. --Tryptofish (talk) 22:52, 18 February 2019 (UTC)
- (For my part, I probably wouldn't have considered commenting here at all if I didn't know the atmosphere was being controlled by the DS! And yes, I meant WT:MEDRS, sorry for the mixup.) Your proposal below is in line with the sort of thing I was thinking of for the ideal case - minority (or fringe) positions, if mentioned, must be placed in the context of the majority view, and any addition would have to comply with that. However, I agree with the comment that this isn't coming across very well for the existing text. In particular, to me the weighting in the current version of the lead seems to imply there is at least a serious possibility of cancer risk: emphasis on "concerns" with no direct counterbalance, a lot of use of attribution and quotation, the 2014 meta-analysis getting the final word in paragraph 4, and IARC being given the most visible position in paragraph 5. The situation becomes clearer after reading the body, but the same kind of thing applies there too, such as giving the "limited evidence" statement the first position in the paragraph that describes the consensus statement. Sunrise (talk) 04:08, 18 February 2019 (UTC)
- Sunrise, I've taken a stab and rearranging this section a little bit to at least be more in line with what an intended reading should be with appropriate weighting. That would be the prime spot for adding the example citations like Tryptofish mentioned. That being said, it's currently written to rely on other studies commenting on the minority viewpoint rather than standalone, so I'm still not sure how best to fit in the new study with regards to my SYNTH comment above. If this version at least makes better sense, we can go forward from there on whether it's better to wait awhile before adding anything about this new study. We'll be getting ourselves into less trouble at least trying to add it in to some variation of the change I just made.
- Something like that could also be copied into the lede too to replace the problem areas you brought up if people are ok with that idea. I'm not beholden to my particular version, but that's the gist to get across at least using the sources we have so far. I can do more digging if we want to discuss specifically how international agencies have discounted many meta-analysis in this subject, though it may be redundant to some degree too. Kingofaces43 (talk) 20:12, 18 February 2019 (UTC)
- I've been thinking about this, and while I would oppose giving this one paper a prominent treatment in the form of something like a paragraph of its own, I do think that we could cite it in the form of something like: "Although some reviews of the literature have concluded that there is evidence of occupational carcinogenicity, (cite papers, including this one) most analyses have concluded that the weight of evidence is against that being the case. (cites)" Or something like that. --Tryptofish (talk) 23:36, 17 February 2019 (UTC)
- I’ve been batting around a similar idea. Basically this conversation so far is showing readers aren’t catching that scientific agencies haven’t been giving these studies weight due to various issues in the studies (that would also at least in part be inherent in this study too). We skim over that in our current content aside from the single sentence mentioning bias and confounding, so that’s a prime candidate to flesh out. Your example could fit in that paragraph, but it also wouldn’t really matter whether this new study is included or not since the sources already mention them. I’ll looking into fleshing out based on previous sources in the morning.
- The only caution I would have against citing the study even as an example would by a SYNTH concern. The sources we can bring up talk about previous meta-analyses and the underlying literature’s issues that this new study would have run into as well, but those sources don’t discuss this new study directly. Not sure how best to tackle that aspect if someone is adamant about including the example listing as you mention quite yet, but your approach looks ok. Kingofaces43 (talk) 01:34, 18 February 2019 (UTC)
- I think we are moving in the right direction, but I'll wait to see what other editors think at this point. --Tryptofish (talk) 22:54, 18 February 2019 (UTC)
Wikipedia is an encyclopedia, which means to "give information on many aspects of a subject." So the high barrier should be against suppressing any significant aspect of the evidence, shouldn't it? Not against including it. Even CNN reported the new meta-study. https://www.cnn.com/2019/02/14/health/us-glyphosate-cancer-study-scli-intl/index.html . On WP we have to be more PC and MSM than the corporate media?
- FWIW The reason I ended up on this page is my neighbors have evidently killed off the bees in our neighborhood this spring by using herbicide on their weeds. No bees, no fruit on my trees. In the book Seeds of Destruction, Engdahl claims Roundup is the cause of the bee die off. Now I believe it. There's primary evidence for you.
- Probably no point posting Engdahl as a source on the main page. It will be whitewashed away faster than Tom Sawyer's fence.
- WP is a fabulous resource, but many people do not trust it on controversial issues, because of this tendency to downplay controversy and focus only on a "consensus" view. By definition consensus is not unanimous. And who decides which controversies are worthy of note? For instance, there is a virtually unanimous consensus that the earth is round. Yet WP has a bundle of pages on Flat Earth theory. Which conceivably are not a threat to any corporate interest.
- To a non Wikipedian like me, it's pretty mind boggling that WP admins can justify referencing a study that supports a corporate viewpoint, yet ban mention of the corporate sponsorship. That's such a basic rule of journalism. How anyone can hold those contradictory views - and then delete a science article from a university website as a puff? This reminds me a bit of East Block totalitarian doublethink, or just plain bias. Sorry if I'm not being polite, I'm actually trying, but sometimes being frank and being polite are contradictory positions too.
- Indeed, WP does often include a paragraph for contrary views. In the case of Global Warming, there is this page: List of scientists who disagree with the scientific consensus on global warming. So I don't see how global warming can be used as an example for suppressing opposing views about glyphosate.
JPLeonard (talk) 10:42, 11 May 2019 (UTC)
- Please keep in mind that editors can be in favor of presenting content in a particular manner, without being motivated by supporting "a corporate viewpoint", and without it being "whitewashing". The editing community went through an epic process at Wikipedia:Arbitration/Requests/Case/Genetically modified organisms over these issues. I agree with you that Wikipedia is rather poor at covering controversial current events, but we would be even worse if we simply went by editor opinion. You may perhaps find WP:RGW interesting reading. --Tryptofish (talk) 18:56, 11 May 2019 (UTC)
- Hi Tryptofish, I am glad we agree on something, "that Wikipedia is rather poor at covering controversial current events". However, I'm not talking about the manner of presentation, but about deleting entirely, about suppression of noteworthy opposing views or data. And I'm not suggesting to go by editor opinion. Just state the facts. Such and such a study said so and so. So we are both in favor of improving WP on that score.
- To me, Flat Earth is completely untenable. According to WP, the Greeks had worked out that it's round by 600 BC, and Eratosthenes even measured its circumference accurately around 200 BC. Nowadays you can call anybody in a different time zone and ask them the time, and work it out yourself. So how do some otherwise rational people still join the Flat Earth Society? From what I've seen, it's echo chambering, cherry picking and purity spiraling.
- What I'm getting at, is that a consensus may be influenced within any group by such factors. So one should be careful to distinguish "consensus" from certainty, because it is in flux all the time, as noted above by others. Without considering objections to the status quo, we don't move forward.
- About the GMO issue, that's closely related to glyphosates, Monsanto, Roundup Ready and F Wm Engdahl's book. I see that there is a controversy section there Genetically_modified_organism#Controversy but apparently the page is locked, is that correct? The template also has no section for books. There is also a section on the right text box for Controversies with 5 entries, starting with Genetically modified food controversies so one can see that there has been effort put into this, which of course reflects the intensity of anti-GMO protests worldwide.
- So it seems that WP does deal with controversies on current events, even closely related ones, and what several posters here are wondering is, why not here? JPLeonard (talk) 19:50, 11 May 2019 (UTC)
- This page is about glyphosate as a specific chemical entity, whereas the controversy section of the GMO page, and the Genetically modified food controversies page, are obviously the best places for controversies. It makes better sense to cover information in the most closely related articles, as opposed to having redundant stuff appearing in multiple pages. If those pages are restricted to editing by longer-term editors, you can always post an edit request on the corresponding article talk page. Wikipedia:Edit requests describes how to do that. This talk section, here, is from February, when editors were discussing whether or not to source the information to a press release, before the actual study came out. If you look lower on this talk page, where the most recent discussions are, you will see me linking to the now-published study, and advocating that we cite it heavily, and other editors largely agreeing with me about that, although we are still discussing how, exactly, to frame it. --Tryptofish (talk) 21:41, 11 May 2019 (UTC)
- I'm confused. A herbicide is not a food so how can the place for this be the GMO food page? Anyway, the article already contains a discussion with opposing views about risks of this chemical:
- "the available data is contradictory and far from being convincing" with regard to correlations between exposure to glyphosate formulations and risk of various cancers, including non-Hodgkin lymphoma (NHL).[6] A meta-analysis published in 2014 identified an increased risk of NHL in workers exposed to glyphosate formulations.[7]"-- JPLeonard (talk) 02:07, 18 May 2019 (UTC)
- Looking back at my earlier comments to you, I'll confess to having been a little confused myself.
This section of the talk page is from a while back, and there have been a lot of different things being discussed more recently, so I feel like it has been a lot for me to juggle. Actually, the meta-analysis is going to be added to this page. It would be better to discuss it down there, in the more recent talk page discussions below. But it's reasonable to consider herbicides that are used on GM food crops to be part of the controversies over GMOs and related technologies. --Tryptofish (talk) 18:03, 18 May 2019 (UTC)
- Looking back at my earlier comments to you, I'll confess to having been a little confused myself.
- "the available data is contradictory and far from being convincing" with regard to correlations between exposure to glyphosate formulations and risk of various cancers, including non-Hodgkin lymphoma (NHL).[6] A meta-analysis published in 2014 identified an increased risk of NHL in workers exposed to glyphosate formulations.[7]"-- JPLeonard (talk) 02:07, 18 May 2019 (UTC)
I'm hesitant to reopen the topic of including the Zhang (2019) paper in this or one of the other articles, but the Washington Post has mentioned it twice July 26, 2019 - Judge cuts award and August 1, 2019 - Merril Hoge. The WaPo articles are missing the criticisms of Zhang (2019) raised here and elsewhere. I understand the need to follow WP:MEDRS, but I also feel that it is important to cover legal cases and media coverage so that readers who look to WP can get more information on a topic. Perhaps there should be a sentence about Zhang (2019) that also cites some of the criticisms? I saw Tryptofish's comment below under "Better Sourcing Needed" about Geoffrey Kabat and Forbes and while I might possibly agree, I do think that there should be something about Zhang (2019) here or elsewhere. Cxbrx (talk) 03:09, 3 August 2019 (UTC)
- Well, I've felt for a while that the sourcing here needs to be updated, and I also feel utterly exhausted over the battles that have been going on here. My suggestion would be to start a new talk section at the bottom of the page, rather than reopening old talk threads, and I'd like to hear from other editors about it. --Tryptofish (talk) 17:40, 3 August 2019 (UTC)
BRD: Should Monsanto-funded studies be marked as such?
There is a discussion going on in edit summaries about whether research funded by Monsanto should be marked as such (see most recently here). This, I suppose, is a sort of intellectual mise en abîme. Wikipedia policy says that contributors having a vested interest must declare that interest on the talk page and wait for other contributors to validate (or reject) their proposed contributions.
The article cited includes a disclosure statement saying that one of the principal authors has worked, as an employee of the subcontractor "Exponent", for Monsanto. The authors are employed by Exponent, a scientific research and consulting firm that provides services for private and governmental clients, including on projects concerning glyphosate and other pesticides. In the past five years, Ellen Chang has provided consulting services through Exponent on behalf of Monsanto Company on other issues, and she also has provided consulting services on other pesticides and lymphohematopoietic cancers for other clients.
It also notes that the research was wholly funded by Monsanto.(source)
I see no reason for en.wiki not to indicate who paid for the research. Doing so does not invalidate the study, but dutifully mirrors the disclosure statement and funding sections in the conclusions to the article. IMO, 19 bytes won't break the servers' backs. SashiRolls t · c 17:09, 5 May 2019 (UTC)
- About a month ago, a statement about funding was added in the "Cancer" section: [5], and I tagged it, with the edit summary
Saying "funded by Monsanto" in this way seems to me to go too far towards editorializing in Wikipedia's voice. Do sources demonstrate that the paper was actually biased as a result?
: [6], which I think is self-explanatory. Actually, despite my most recent edit summary, I misremembered what I had done as a revert; in fact, all I did was put a tag after the phrase, explaining the tag in my edit summary just above, and the tagged passage is still there as of today. About a month later, the same statement about the same source was added in the "Human" section, without taking note of the earlier tag: [7]. And that is what I reverted yesterday: [8].
- As per my earlier edit summary, saying "funded by Monsanto" in Wikipedia's voice makes it sound like Wikipedia is passing judgment on the soundness of the study, which violates WP:NOR. The correct way to write about it would be to cite a reliable source, such as a journalist or an academic (and not a conspiracy theorist), writing about the funding of the study and saying that the funding had a noteworthy influence on what the study said – and to attribute that view to that source. And if there are different sources that come to differing conclusions about it, we would have to reflect that difference. But not to say it in Wikipedia's voice. If we can say it in an NPOV way, with proper attribution, that would be fine with me. I'm not looking to prevent the information from appearing in an appropriate way. (It certainly has nothing to do with the number of bytes.)
- I have a concern about the quoting of the WP:COI policy. If it was simply intended to draw a parallel between COI issues at Wikipedia, and possible influences of funding on the source being discussed, that's fine, no problem. But if there were any intent to imply that I am an editor with a COI and should not have made the edits that I did, that's a problem. --Tryptofish (talk) 21:29, 5 May 2019 (UTC)
- Sources too closely associated with the subject, can prevent an article from being neutral. "Articles should be based on reliable, third-party, published sources with a reputation for fact-checking and accuracy." (WP:RS/overview) When the author is writing about the toxicity of a product that is sold by his employer he may have a conflict of interest. JimRenge (talk) 22:37, 5 May 2019 (UTC)
- This does seem straightforward. Without the declaration of funding the paper would not have been published in the source (Journal of Environmental Science and Health). Why would different standards hold here? SashiRolls t · c 04:17, 6 May 2019 (UTC)
- At the same time, the journal did decide to publish the paper, calling it a "systematic review" rather than an opinion piece, and it would be original research for Wikipedia to imply in Wikipedia's voice that there had been something dodgy about the editorial decision to publish it that way. Our standards require us to go by what the sources say, not by what editors think about the sources (and because this is health-related, our requirements for going with what academic journals decide are reviews are all the more stringent). But as I already said, I would be happy to support adding the information, if it were attributed to secondary, reliable sources that say that the funding is something that, in the opinion of the secondary sources, is important to consider when evaluating the paper. --Tryptofish (talk) 17:18, 6 May 2019 (UTC)
- Certainly considering that it is extremely unlikely that any corporation has ever published a study or a review that found their product to be problematic it should not be considered unimportant to know exactly just who paid for the study or review. If we are not allowed to mention funding I assume we have a guideline saying just exactly that. Right? Gandydancer (talk) 18:01, 6 May 2019 (UTC)
- Who is the editor who said that we should never mention it? Certainly not me.
- (Parenthetically, one of the most landmark analyses of climate change was written by a scientist who accepted funding for it from the Koch brothers and then used their funding to publish the opposite of what they wanted. And I have a friend who took funding from the tobacco industry to support studies he published that demonstrated the toxicity of nicotine. Scientists actually do that sort of thing a lot more often than the public realizes. And anyway, Monsanto did not publish the review. A peer-reviewed scientific journal did.) --Tryptofish (talk) 18:08, 6 May 2019 (UTC)
- Certainly considering that it is extremely unlikely that any corporation has ever published a study or a review that found their product to be problematic it should not be considered unimportant to know exactly just who paid for the study or review. If we are not allowed to mention funding I assume we have a guideline saying just exactly that. Right? Gandydancer (talk) 18:01, 6 May 2019 (UTC)
- At the same time, the journal did decide to publish the paper, calling it a "systematic review" rather than an opinion piece, and it would be original research for Wikipedia to imply in Wikipedia's voice that there had been something dodgy about the editorial decision to publish it that way. Our standards require us to go by what the sources say, not by what editors think about the sources (and because this is health-related, our requirements for going with what academic journals decide are reviews are all the more stringent). But as I already said, I would be happy to support adding the information, if it were attributed to secondary, reliable sources that say that the funding is something that, in the opinion of the secondary sources, is important to consider when evaluating the paper. --Tryptofish (talk) 17:18, 6 May 2019 (UTC)
- I fixed it myself: [9]. And it wasn't that difficult to do. --Tryptofish (talk) 18:46, 6 May 2019 (UTC)
- P.S. I also noticed that some more recent reviews have cited the 2016 one. On a quick read, two agreed with the 2016 review and one disagreed; I haven't looked into the authors or their funding. I don't have time right now, but it would be worth looking into that further and maybe updating the page accordingly. --Tryptofish (talk) 18:55, 6 May 2019 (UTC)
- Tryptofish, I'm just catching up, but where did this NRDC source come in? Had it been discussed before and I missed it, or was it just brought in? The issue I'm seeing is that it is not MEDRS (not to mention an advocacy group) and can't really be used to criticize a MEDRS source.
- Below, I mentioned two reviews, but what's the third review you mention? At least in my look of the secondary sources so far barring a third one, I'm not seeing any criticism of the 2016 review we can use here. Kingofaces43 (talk) 23:06, 6 May 2019 (UTC)
- You asked me just as I was about to log out for tonight, so I'll have to get back to you tomorrow about the reviews that I saw. The NRDC came from me, in my comment immediately above, about "I fixed it myself". I'm conflicted about this: I at least partly agree with you about MEDRS issues; on the other hand, I appreciate that there is a controversy that is separate from the health science aspects, and I don't want to leave that out, and although NRDC is an advocacy group, I think they are reasonably mainstream. --Tryptofish (talk) 23:13, 6 May 2019 (UTC)
- OK, I've gone back and looked closely at sources. First of all, I have some hesitation about the two that you listed just below. Both of them only mention the 2016 review very briefly, in passing, without really expressing an evaluation of it one way or the other, more like "here are some analyses that found a correlation, and here are some that didn't", with Chang 2016 among the latter. And the second one has a lot of authorship from Monsanto people, so it will be more of a battle to get editors to agree to using it than it's worth, given how little it really tells us about Chang 2016.
- The three that I saw were:
- This, which on looking more closely I am going to quickly rule out, because it's on Beall's List of predatory journals.
- A large health study from 2017. It cites Chang favorably and reaches the same conclusions (albeit with some caveats about not-quite significant effects in "highest exposure" groups). It's not exactly a review article though. The authors seem to have no conflicts of interest.
- A meta-review from 2019. Cites Chang, no COI issues. But it concludes that there is a "compelling link" between glyphosate and those kinds of cancers.
- I think we might want to significantly update the sourcing we use on this and related pages, but we must include that meta-review (#3) and present both sides. --Tryptofish (talk) 20:20, 7 May 2019 (UTC)
- On the two I cited, I was not saying they were comprehensive, but that of reviews citing it, nothing negative came up (as opposed to glowing in-depth reviews which I did not say). Hopefully that makes it a little more clear. Kingofaces43 (talk) 04:02, 8 May 2019 (UTC)
- In general, it's inappropriate to call out funding source alone on peer-reviewed research when it's done by independent researchers, and a common mistake made by non-scientists as Tryptofish alluded to above. If there was some connection with the funding source that biased the study design or findings, then peer-reviewers or future MEDRS sources will ultimately call that aspect out, not us. If I'm peer-reviewing a paper for a journal that has an article with industry funding, it gets treated the same as any other where I try to fish out methodological issues, etc., though maybe with more scrutiny on methodological "positive" biases. If it passes peer-review, we can't cast shade on an article for solely due to funding source done by independent researchers.
- In this case, the disclosure statement makes clear the authors had sole control of the article, not the funder. That's the main criteria for independence for us here (if a current Monsanto employee were an author, that would be a different story). Independent consulting agencies alone are more variable quality quality compared to university researchers, but while Chang has the affiliation with the independent consulting agency, both Chang and Delzell are affiliated with Standford University.
- I took a look on Web of Science though to see what MEDRS reviews are citing the study. There are two:
- Re-registration Challenges of Glyphosate in the European Union
- Glyphosate epidemiology expert panel review: a weight of evidence systematic review of the relationship between glyphosate exposure and non-Hodgkin’s lymphoma or multiple myeloma
- In either case, there is no criticism of methodology, findings, etc. It's been nearly three years with no critical coverage so we have to consider criticisms of the meta-analysis with respect to just funding source WP:UNDUE unless a source shows there was a problem with the funding source. Kingofaces43 (talk) 22:57, 6 May 2019 (UTC)
- Nobody is "shading on an article". The reverted addition was a neutral and verifiable statement of fact: "funded by Monsanto". Wikipedia should not have lower standards for documenting that fact than the original journal does. The idea that the original journal is committing "a common mistake made by non-scientists" by reporting the funding is strange. SashiRolls t · c 05:46, 7 May 2019 (UTC)
- SashiRolls’ reasoning seems clear and correct. The reverted addition should be restored. Jusdafax (talk) 06:29, 7 May 2019 (UTC)
- What's wrong with what I put in about it? --Tryptofish (talk) 18:48, 7 May 2019 (UTC)
- In case editors did not see it: [10]. --Tryptofish (talk) 21:14, 7 May 2019 (UTC)
- It seems ok, but let's look to be sure. Hmmm... the edit summary (§) doesn't get a very good "plays well with the other children" grade (0/1). Hmm... it does seem to be an effort to sidestep the ethical question of how the source should be represented in the first place (0/2), but it also made clear -- at least to me -- that the "dose makes the poison" hatnote needed reverting over at Roundup (1/3) and it seems like a reasonable compromise locally, as long as we reject categorically the premise that a secondary source is necessary to transcribe funding info from an original source (2/4). It was also probably time to remove the call (as you did) to add stuff to that entry from fr.wp. (so 3/5) SashiRolls t · c 21:39, 7 May 2019 (UTC)
- Another editor had said that the removed material should be put back, so I wanted to make sure that editors were aware that the material was, in fact, put back – but in a form that follows policy. --Tryptofish (talk) 21:54, 7 May 2019 (UTC)
- Just out of curiosity, which policy? A primary source showing that the study was funded by Monsanto would be a bank statement showing the transaction, or a form filled out by a contributor saying their research had been so funded. The Journal of Environmental Science and Health is a secondary source for such a claim, unless I'm mistaken in my reading of wp:secondary. SashiRolls t · c 22:04, 7 May 2019 (UTC)
- Another editor had said that the removed material should be put back, so I wanted to make sure that editors were aware that the material was, in fact, put back – but in a form that follows policy. --Tryptofish (talk) 21:54, 7 May 2019 (UTC)
- It seems ok, but let's look to be sure. Hmmm... the edit summary (§) doesn't get a very good "plays well with the other children" grade (0/1). Hmm... it does seem to be an effort to sidestep the ethical question of how the source should be represented in the first place (0/2), but it also made clear -- at least to me -- that the "dose makes the poison" hatnote needed reverting over at Roundup (1/3) and it seems like a reasonable compromise locally, as long as we reject categorically the premise that a secondary source is necessary to transcribe funding info from an original source (2/4). It was also probably time to remove the call (as you did) to add stuff to that entry from fr.wp. (so 3/5) SashiRolls t · c 21:39, 7 May 2019 (UTC)
- SashiRolls, it's best not to misrepresent my comments like that. No one was saying the journal made any "mistakes" It's standard practice to include funding sources in the acknowledgments. It's not standard practice mention funding source for every study cited on Wikipedia though. Either there's a legitimate reason why funding source is an important detail outlined in MEDRS sources for criticism of the review, or there's nothing particularly relevant about the funding source. Kingofaces43 (talk) 05:02, 8 May 2019 (UTC)
- While I may have misunderstood your comment, I don't believe I have misrepresented it:
- In general, it's inappropriate to call out funding source alone on peer-reviewed research when it's done by independent researchers, and a common mistake made by non-scientists as Tryptofish alluded to above.
- The disclosure statement & funding sections of the published article are separate from the acknowledgments section. Those who want to discuss the matter of who is doing the "calling out" can reach those two sections quickly by using this internal anchor provided to "conclusions" in the initial publication being used as a source: (#_i40). Frankly, I have no opinion on the "truth" of the matter concerning gylphosate (I am not competent to judge). However, judging is not a volunteer contributor's task, though NPOV is. And NPOV would seem to suggest that we should follow the "best practices" of the journal we are citing. This doesn't seem that complicated, biased, or even worthy of great walls of text. What gives? SashiRolls t · c 16:00, 8 May 2019 (UTC)
- The fact that editors should not judge in that manner is exactly why WP:NPOV calls for biased statements to be attributed rather than asserted in Wikipedia's voice. Wikipedia policies and guidelines are the "best practices" that Wikipedia should follow. And complying with that is indeed what editors are responsible for, which is why I revised the way that the page refers to the funding. Now that it has been fixed, I think that "what gives" is it's time to give this wall of text a rest. --Tryptofish (talk) 16:54, 8 May 2019 (UTC)
- I for one am willing to let your new text be for now with the MEDRS tag in place rather than removing the source outright for now. It doesn't look like anything else is going to get resolved here and now at least. Kingofaces43 (talk) 23:59, 8 May 2019 (UTC)
- The Revised Glyphosate Issue Paper: Evaluation of Carcinogenic Potential. EPA's Office of Pesticide Programs. December 12, 2017. page 22, introduces the Chang/Delzell paper as follows: "This list was also cross-referenced with review articles from the open literature [Chang and Delzell (2016), Greim et al. (2015), Kier and Kirkland (2013), Kier (2015), Mink et al. (2012), Schinasi and Leon (2014), and Williams et al. (2000)] Footnote 11: All review articles, except Schinasi and Leon (2014), were funded and/or linked to Monsanto Co. or other registrants." It seems that EPA thought the readers of their study should be informed about the funding of these reviews. JimRenge (talk) 16:23, 10 May 2019 (UTC)
- A footnote like that falls into passing mention. Is there any significant discussion of funding source with respect to validity of the review? That's ultimately what we should be shooting for. Kingofaces43 (talk) 18:43, 10 May 2019 (UTC)
- The Revised Glyphosate Issue Paper: Evaluation of Carcinogenic Potential. EPA's Office of Pesticide Programs. December 12, 2017. page 22, introduces the Chang/Delzell paper as follows: "This list was also cross-referenced with review articles from the open literature [Chang and Delzell (2016), Greim et al. (2015), Kier and Kirkland (2013), Kier (2015), Mink et al. (2012), Schinasi and Leon (2014), and Williams et al. (2000)] Footnote 11: All review articles, except Schinasi and Leon (2014), were funded and/or linked to Monsanto Co. or other registrants." It seems that EPA thought the readers of their study should be informed about the funding of these reviews. JimRenge (talk) 16:23, 10 May 2019 (UTC)
- I for one am willing to let your new text be for now with the MEDRS tag in place rather than removing the source outright for now. It doesn't look like anything else is going to get resolved here and now at least. Kingofaces43 (talk) 23:59, 8 May 2019 (UTC)
- The fact that editors should not judge in that manner is exactly why WP:NPOV calls for biased statements to be attributed rather than asserted in Wikipedia's voice. Wikipedia policies and guidelines are the "best practices" that Wikipedia should follow. And complying with that is indeed what editors are responsible for, which is why I revised the way that the page refers to the funding. Now that it has been fixed, I think that "what gives" is it's time to give this wall of text a rest. --Tryptofish (talk) 16:54, 8 May 2019 (UTC)
- While I may have misunderstood your comment, I don't believe I have misrepresented it:
- Yes, of course we should tell the reader that Monsanto/Bayer is funding research intended to benefit the company. I'm not going to argue the idealized notion that a peer-reviewed publication always and completely clears the source study of bias. Instead, I note that there have been many articles written about how Monsanto/Bayer has been and still is funding research about the company's products, and how that research is being questioned for apparent bias. The idealized case is not what we have here – instead, this is a matter of following the reliable sources in their questioning of the funding sources. Binksternet (talk) 21:11, 7 May 2019 (UTC)
I'm not going to argue the idealized notion that a peer-reviewed publication always and completely clears the source study of bias.
No one ever advanced that idea in the first place (though it can be a common mistake). The point was that the first step of assessment is peer-review where reviewers look for biases, methodological issues, etc. If a problem did get missed in a MEDRS source like that, we wait for criticism from other MEDRS sources. So far, we don't have anything like that. Kingofaces43 (talk) 05:02, 8 May 2019 (UTC)
Trypto has again reverted the inert statement that the 2016 meta-study is funded by Monsanto, preferring instead to suggest it's people he considers to be wacky who point that out. As Jim Renge pointed out above, the EPA itself makes this clear, as I pointed out the original pub makes this clear. If you've read the 2019 Benbrook article SmartSE suggests including, this is a major aspect of the difference between the EPA and the IARC evaluations (along with the fact that the EPA seems to have studies only on technical glyphosate (i.e. unmixed with additives that let it get under the skin more readily, with AMPA, etc.)). So far, jusdafax, Gandydancer, Binksternet, Jim Renge and I seem to think it's no big deal to make brief mention of the fact without fanfare, whereas Trypto wants to write an entire sentence to add new information rather than just attribute the statement. Kingofaces43 wants to go farther, adding a tag of shame to that sentence for maximal rhetorical effect. Right now I see a 5-2 split in favor of straightforward reporting the question, as Gandy had done, and as I did here. Please if I'm mistaken in my reading of your opinions above, feel free to let me know. Why are you editing against what seems to be a pretty strong 5-2 consensus Tryptofish? SashiRolls t · c 15:59, 13 May 2019 (UTC)
- It's not the way that you are framing it. I'm adhering to Wikipedia:Neutral point of view#Attributing and specifying biased statements. If you think that I'm being disruptive, WP:AE is that-a-way. --Tryptofish (talk) 16:12, 13 May 2019 (UTC)
Section break (BRD)
Well here we go again. Pinging JimRenge, SashiRolls, Binksternet, Tryptofish, and Jusdafax Almost six months ago KoA was the only editor with an objection to either noting the fact that Monsanto funded this favorable review or including a RS that noted that Monsanto paid for it as was suggested by Trypto who said:
- The correct way to write about it would be to cite a reliable source, such as a journalist or an academic (and not a conspiracy theorist), writing about the funding of the study and saying that the funding had a noteworthy influence on what the study said – and to attribute that view to that source.
At that time KoA tagged it and said: "I for one am willing to let your new text be for now with the MEDRS tag in place rather than removing the source outright for now."
Almost six months went by and KoA removed it saying:
- Been tagged for awhile, and still not seeing any WP:MEDRS sources criticizing the study itself at last check.
I responded by adding the previous edit that stated that the study was funded by Monsanto which KoA promptly removed saying:
- Not appropriate to restore content specifically flagged as poorly sourced, especially while removing the flag. There explicitly was not consensus on the talk page for this. Again, please find appropriate WP:MEDRS sources as discussed on the talk page and follow WP:ONUS policy.
In truth, there was agreement to go with Tripto's solution with only KoA objecting. Is it now OK to flag something even though you are alone in your decision that certain copy needs to be flagged and then come back six months later and say Well no one has decided that I was not right so I guess I must be right so I'm going to remove this copy? Gandydancer (talk) 19:25, 27 September 2019 (UTC)
- Per WP:BIAS, potentially biased sources may make in-text attribution appropriate. I think the funding of the study by Monsanto makes it potentially biased and I support Gandydancer´s edit. JimRenge (talk) 20:58, 27 September 2019 (UTC)
- Let me start by saying that I don't understand what JimRenge just said, because Gandydancer's edit had the effect of saying it without in-text attribution.
- So, this began with KofA making this edit: [11]. Let's be clear about what I did or did not support before: what I support is what KofA removed in that edit (minus the MEDRS tag). I disagree that this particular passage is really something that falls under MEDRS, because no MEDRS claim is being made, just a claim about funding. The next edit, by Gandydancer, was this: [12]. I do not support that edit, and there is no consensus for it, because it makes the statement in Wikipedia's voice, without attribution or sourcing. Then there was some edit warring, leading to this most recent (as of when I write this) edit: [13]. I disagree that such talk page consensus exists. Rather, I think that the closest thing to consensus was what we had prior to the edits today, but minus the tag. --Tryptofish (talk) 22:27, 27 September 2019 (UTC)
- Sorry, looking back at what I said earlier, what I actually supported is having that statement with a better source. But I don't support simply removing the statement; I'd prefer substituting a better source. And I sure do not support saying it in Wikipedia's voice with no attribution. --Tryptofish (talk) 22:33, 27 September 2019 (UTC)
- The in-text attribution is [14], p. 424. JimRenge (talk) 23:15, 27 September 2019 (UTC)
- So you are talking about Chang and Delzell, 2016. Yes, that is the source at the end of the sentence. And yes, it provides reliable sourcing for "According to a systematic review and meta-analysis published in 2016, when weak statistical associations with cancer have been found, such observations have been attributed to bias and confounding in correlational studies due to workers often being exposed to other known carcinogens." No problem there. But that is not the issue here. The issue is the insertion of the words "funded by Monsanto" into that sentence, as in this edit: [15]. I just re-read the source. Unless I'm missing something, and I don't think that I am, although the authors go into a lot of detail about possible sources of bias in carcinogenicity studies of glyphosate, they do not discuss the role of funding of studies in those biases. If I missed it, please show me where it is. Consequently, it does not constitute attribution for making a clearly POV statement in Wikipedia's voice. --Tryptofish (talk) 18:53, 28 September 2019 (UTC)
- The in-text attribution is [14], p. 424. JimRenge (talk) 23:15, 27 September 2019 (UTC)
- Just a note... While I do believe that it is not only proper but in fact should be required to mention when a review is done by the very corporation that sells the product of the review I do, regrettably, after many years of dealing with the politics of this place, come hat-in-hand to say the the Trypto edit which includes the Sass comments is acceptable to me. In fact I welcome it and considering the alternative am grateful... Gandydancer (talk) 21:45, 29 September 2019 (UTC)
- Thanks, I do understand how difficult these issues are, and I genuinely appreciate your willingness to accept that edit. I hope that others will accept it, too, and that we can now move on. --Tryptofish (talk) 23:17, 29 September 2019 (UTC)
- Thanks as well and I genuinely accept and appreciate your willingness to work as a team willing to meet in the middle when that is what is needed to do. I hope to move on as well, though I see no reason to tag your suggested edit since there is no need for that. Gandydancer (talk) 23:31, 29 September 2019 (UTC)
- All good, and I agree that there is no justification for tagging the source as non-MEDRS. If anyone disagrees about that, I'd rather start an RfC than continue to argue about it. --Tryptofish (talk) 20:39, 30 September 2019 (UTC)
- Thanks as well and I genuinely accept and appreciate your willingness to work as a team willing to meet in the middle when that is what is needed to do. I hope to move on as well, though I see no reason to tag your suggested edit since there is no need for that. Gandydancer (talk) 23:31, 29 September 2019 (UTC)
- Thanks, I do understand how difficult these issues are, and I genuinely appreciate your willingness to accept that edit. I hope that others will accept it, too, and that we can now move on. --Tryptofish (talk) 23:17, 29 September 2019 (UTC)
- Just getting back from a weekend trip, so I'm caching up since nothing was happening with this when I left. I'll start here though. In short, the core problem in the related edits was using funding source to discredit the findings of a MEDRS study. There's really no other reason why that would be pointed out here and not for every other source out there, and I really love to hear a valid reason that doesn't step into that territory at all. That's where we get into trouble with MEDRS. That WP:BURDEN wasn't even on me for finding such sources in order to remove the content, but I went looking pretty hard last week and found nothing that would support it. I thought I was being pretty fair in terms of WP:DEADLINE too, it seems it's too soon.
- Just a note... While I do believe that it is not only proper but in fact should be required to mention when a review is done by the very corporation that sells the product of the review I do, regrettably, after many years of dealing with the politics of this place, come hat-in-hand to say the the Trypto edit which includes the Sass comments is acceptable to me. In fact I welcome it and considering the alternative am grateful... Gandydancer (talk) 21:45, 29 September 2019 (UTC)
- In short, I definitely at least agree on things you say we definitely shouldn't be doing or needing MEDRS for more in-depth discussion, and you should not have needed to go through the effort of rehashing again why we definitely don't put something like this straight in Wikipedia's voice. Your additions help, but we also haven't quite got to the point where WP:DUE even with the attribution is really established with the NRDC source. If Sass was publishing in even say an editorial in a journal, I'd be more comfortable with the weight aspect than coming straight from the NRDC site, but that basically means the talk page is unchanged since May at this point.
- I was going to say I'm fine leaving this at the old talk page status quo of reminding editors who strongly want to include something on this to find more appropriate sources and leave the article as-is for now. However, am I missing something in terms of what the current text is based on? I just went back to the source, and this is the only time Sass mentions the paper in question really:
Also published since IARC’s assessment is a 2016 Monsanto-sponsored systematic review and meta-analysis of epidemiologic studies, including the Ag Health Study, which specifically identifies a statistically significant risk of NHL from glyphosate exposure (Chang and Delzell 2016), strengthening the IARC assessment conclusions from a year earlier.
[16] along with a namedrop in the next paragraph. There isn't any criticism of this study or really any discussion of it all to match the text. That's the only thing that threw a wrench into ending this comment on the first sentence of this paragraph. Kingofaces43 (talk) 04:03, 1 October 2019 (UTC)
- I was going to say I'm fine leaving this at the old talk page status quo of reminding editors who strongly want to include something on this to find more appropriate sources and leave the article as-is for now. However, am I missing something in terms of what the current text is based on? I just went back to the source, and this is the only time Sass mentions the paper in question really:
- About what Sass says in that passage, she is saying that she interprets Chang and Delzell's material to support a link between glyphosate and NHL, which is is pretty much the opposite of what Chang and Delzell say, and her inclusion (in her voice, of course) of the mention of sponsorship is clearly, in context, intended as a criticism. And since you say that I "should not have needed to go through the effort of rehashing again why" my position is what it is, I'm going to say that I should not have to do that for you either. I hope you'll drop this now. If not, let's have an RfC, because I'm worn out from arguing this stuff. --Tryptofish (talk) 21:28, 1 October 2019 (UTC)
- I asked because nothing on my question had been hashed here yet much less rehashed, but no, the source does not say that at all. I know the topic can be frustrating, but getting these details right is also serious enough I have to bring it up. That's why I'm glad you sorted through what you did above when you should not have had to, but this fell through the cracks. Sass only repeats the finding of Chang and Delzell that a statistical significance was found, but omits the rest of the findings behind that. They never went that step further to criticize anything in that particular paper.
- Speaking broadly rather than just to Tryptofish, the main problem with respect to the content we currently have is that it violates WP:V policy, and I'm more concerned than I was before due to this closer look at the source. Such policy violations by definition cannot gain consensus on a talk page, so there's nothing for me to drop on that subject as anyone is free to remove it at any time due to that. The other is that advocacy non-MEDRS sources, as we've flagged this one as in talk discussion, are prone to cherry-picking, so focusing on what they do include is largely moot. While I've made it clear I'm willing to wait for those that feel strongly to find appropriate MEDRS sources, I'm still going to be consistent in policy and guideline here. If it's another six months, a year, etc. down the road and no one has found a source, no one can complain from a WP:PAG perspective about removal if it has to come to that.
- About what Sass says in that passage, she is saying that she interprets Chang and Delzell's material to support a link between glyphosate and NHL, which is is pretty much the opposite of what Chang and Delzell say, and her inclusion (in her voice, of course) of the mention of sponsorship is clearly, in context, intended as a criticism. And since you say that I "should not have needed to go through the effort of rehashing again why" my position is what it is, I'm going to say that I should not have to do that for you either. I hope you'll drop this now. If not, let's have an RfC, because I'm worn out from arguing this stuff. --Tryptofish (talk) 21:28, 1 October 2019 (UTC)
- The solution has always been extremely simple for someone that wants to tackle this down the line; find a MEDRS source criticizing the study. If none exists detailing criticism associated with funding source, that's telling it's not WP:DUE. If one does exist, then by all means bring it here so content can be fleshed out because it then should be so. I'm willing to wait yet again. Kingofaces43 (talk) 01:20, 2 October 2019 (UTC)
- This has nothing to do with waiting. It is about whether or not a MEDRS source is required here. I'm convinced that you are wrong about this, and you will be just as wrong six months or a year from now. If you continue your misguided demand for a MEDRS source, I will want an RfC. --Tryptofish (talk) 20:37, 2 October 2019 (UTC)
- It's essentially a standard practice with MEDRS sources that you need another MEDRS source to criticize them. Even expert opinions don't always get used carte blanche in that regard when they're of even higher quality (i.e., not self-published by the advocacy group). Otherwise, I can think of a lot of advocacy websites we've not used in the past that would now be opened up if such a change in direction is made. I'm just being consistent across the board with policy and guidelines since I don't want to see that can of worms opened even though it might be tempting for some subject material. I've lost track of how many times I've seen a good writeup by a scientist only to reject my using it here as "expert opinion" because it was through an advocacy group and not a reputable scientific organization.
- The MEDRS issue isn't the main concern now though. The current content just isn't in the source with respect to Sass criticizing this particular study. There is a statement of a study finding and funding source. Nothing more with respect to the study in question, so the current version violated WP:SYNTH too. If it were other areas of discussion in the source, then criticize is appropriate. At best, we can summarize it as
Writing for the Natural Resources Defense Council environmental advocacy group, Jennifer Sass restated the statistical associations found in the review while noting that it was funded by Monsanto.
That at least satisfies the accuracy part for verifiability policy, so I've gone ahead and made the change instead of adding a tag or deleting (and do realize I've been giving a lot to "meet in the middle" even before this). It would be a mess to try pull anything more out of the source for this, and I don't think anyone has the appetite for that. Even that still doesn't satisfy WP:DUE due to it being an advocacy source known for loaded language, but as I've said before, I'm willing to wait in terms of WP:ONUS for more appropriate sources and leave this be at this point. Kingofaces43 (talk) 16:31, 4 October 2019 (UTC)
- This has nothing to do with waiting. It is about whether or not a MEDRS source is required here. I'm convinced that you are wrong about this, and you will be just as wrong six months or a year from now. If you continue your misguided demand for a MEDRS source, I will want an RfC. --Tryptofish (talk) 20:37, 2 October 2019 (UTC)
Proposal: Rewriting of the lead
I've been looking at the article for a while and have read all the discussions currently in this talk page (and some archived discussions that were linked), mostly because I'm working on improving the Portuguese article. And this one is in much better shape than that one, so I often just translate stuff from here to there. But I see a few things that could be improved here, one of them being the lead. It's too long, goes a bit too much into detail, and instead of summarizing the most important content, it has the exact same text as is found elsewhere, making the article a bit too repetitive (add that to the fact that there is also text that is repeated in both the "Glyphosate alone" and "Glyphosate formulations" subsections of "Toxicity", for instance, and the repetitiveness gets too much - but that's not the point of this).
I think it should be reduced to four short paragraphs:
- What it is and what it's used for. Who discovered it and when (can be mostly maintained as is).
- How it works (I think this also can be mostly kept as the third paragraph currently stands, maybe a bit shorter).
- Toxicity/Carcinogenicity: The wide consensus on its safeness, with a short list of some of the agencies that have concluded this without going into too much detail - leave that for its relevant section. A note of IARC's position, since it has been so widely reported, clarifying that it's the minority view.
- Other controversies: concerns about increased use, lawsuits, perhaps others that I can't think of right now.
I think the second paragraph should be trimmed and merged into the first. The little bit about how it works that is currently in the first paragraph should me moved to the "How it works" paragraph (currently the third, I suggest it becomes the second) and that too can by further summarized. But if no one else thinks these are good suggestions, I won't argue.
My main issue is with what currently are the last two paragraphs. Apart from just being too long for a lead, as it currently stands it looks a bit like a false debate. I get the feeling of an undue interpretation of NPOV as if it should be "here are these different positions, make up your own mind about it". Especially given IARC's view is presented first with all the others coming afterwards in this somewhat long paragraph, many readers will just read that and skip the rest. I submit that this gives it undue weight, even though their report is borderline FRINGE. IARC's report has been more widely reported by the media than anything else, but that can be attributed to the fact the "the most widely used herbicide in the world has, once again, been found to be safe" doesn't make such a good headline. This is getting too long already, so I'll wait for some feedback to see if I'm on to something or if I am way off, and maybe we can collaborate. VdSV9•♫ 15:28, 15 October 2019 (UTC)
- There are cherrypicking concerns in the lead, so your framework looks like it would cut to the point better. Part of the problems are the history of this page when the splits between glyphosate alone and formulations happened around the time the ArbCom had to step in, and this article hasn't been significantly rewritten since. Then we had the splits to Glyphosate-based herbicides and Roundup (herbicide) that created even more redundancy outside of the areas within this article you mentioned. This is probably the best article to work on so the others follow it though.
- For the last two paragraphs, you're right that they don't match up with the body of the article. Over at the Roundup page, the carcinogenicity section has a better ordered weighting that's more in line with what you're looking for, but the content also varies across the pages a little too. On this page Glyphosate#Cancer has the broadest overview, though the third paragraph is better left out considering the methodological concerns the scientific community has voiced with that review (i.e., lumping low quality studies together with the Agricultural Health Study data with negative findings to get a significant effect). Still trying to figure out the best way to address that particular study, but it doesn't change the overall weight of evidence of non-carcinogenicity either for lead content. Kingofaces43 (talk) 20:13, 15 October 2019 (UTC)
research funding
[17] This was just in the news, but the info doesn't seem to be that recent, so maybe it's already reflected in the article. I don't really follow this topic so I'm posting the link here in case someone wants to use it. 2601:648:8202:96B0:54D9:2ABB:1EDB:CEE3 (talk) 05:25, 15 March 2020 (UTC)
- As far as I can see, these sources aren't used in the article, so wouldn't affect us in that regard. I had this on my radar with a source I have to dig up again that talked about if the journal was going to do anything not long after this source. However, even if the papers are amended or retracted for not including a study sponsor (the "secret funding" bit in the source headline looks like embellishment without additional details), I'm not sure that a paper simply being about glyphosate would reach WP:DUE here unless it had wider implications like study sponsor actually affecting the findings. I'll keep an eye out for developments though. Kingofaces43 (talk) 06:02, 19 March 2020 (UTC)
- Nevermind, what I thought I read elsewhere was actually just way at the bottom of the source. The journal editor said
The authors did not advise me of the source of the funding. For this reason it was not disclosed. Conflict of interests is important but not relevant in this case. The papers will not be amended or retracted.
. Basically, the authors messed up on acknowledgements, but it's not a conflict of interest in either case. Kingofaces43 (talk) 06:15, 19 March 2020 (UTC) - The authors messed up and there is no reason to be concerned about funding bias. Gandydancer (talk) 14:56, 19 March 2020 (UTC)
- As you're aware of from past content discussions, funding source alone isn't a valid criticism without sources actually pointing out the study was biased due to that in some fashion (i.e., the sponsor being involved in experimental design, which is typically assessed during peer-review). That or just have secondary literature commenting on problems in a study that aren't even related to funding source. That also means that studies funded by the organic lobby aren't outright criticized either solely due to the funding if it's an independent researcher (though an individual would get appropriate heat in a BLP for actively pursuing funding from the industry group promising positive results).
- Funding is unfortunately a vague aspersion often used by fringe-proponent sources in this subject as general hand-waving regardless of study validity (e.g., "GMO studies are all bought off by Monsanto"), so we do need to be more cautious of that subject in this area. In this case, we have a newspaper (generally not reliable for scientific discussion anyways) with headlines very loosely tied to what actually happened while the journal editor themselves saying nothing in those particular studies would have resulted in significant bias or conflict of interest. There isn't really much more to discuss content-wise after seeing the journal editor comment unless there are new developments. Kingofaces43 (talk) 17:04, 19 March 2020 (UTC)
- Certainly, feel free to believe whatever you want to believe. I find the fact that anyone would say oh they just "messed up" when they did not include funding laughable. Gandydancer (talk) 21:01, 20 March 2020 (UTC)
- Yes, it's rather silly to forget funders in your acknowledgements section, but I think you'd have a hard time finding a researcher that either hasn't done that or came very close to it at least once, especially with partially funded projects. In the end, it "just" means not everyone got the credit they should. As the journal editor alluded to, that's very different than any sort of conflict of interest issue, which would be much more serious if such an issue wasn't declared.
- We're kind of getting into WP:NOTFORUM territory at this point, but it's also not uncommon for people to confound the acknowledgements section for a conflict of interest statement. Usually, that is handled in separate sections both in the the paper or at least in the submission process. That's how you get papers that say they were funded by X, X had no role in project development/writing by the independent researcher, and text on no conflict of interest all in the same paper. That's the minimum standard us scientists look for during peer-review with any funding source, and deviation from that where a sponsor somehow having final say on project or editorial oversight is no laughing matter in the scientific community when it comes to rejecting manuscripts. Kingofaces43 (talk) 22:31, 20 March 2020 (UTC)
- Certainly, feel free to believe whatever you want to believe. I find the fact that anyone would say oh they just "messed up" when they did not include funding laughable. Gandydancer (talk) 21:01, 20 March 2020 (UTC)
- Nevermind, what I thought I read elsewhere was actually just way at the bottom of the source. The journal editor said
"Monsanto Papers"
I've seen that there has been a lot of reverting lately, and I hope that we can instead discuss some of those things. I see that editors disagree about having a section header titled "Monsanto Papers" (with a capital P). I'd like to have a clear understanding of how that phrase comes from the sources, and how it relates to WP:NPOV. On the face of it, it sounds to me like the Pentagon Papers, and that implies in Wikipedia's voice that there is a predominant view in reliable sources (not advocacy groups) that this has been a scandal on the order of the Séralini affair. That's making me a little nervous about using that particular section header. --Tryptofish (talk) 23:34, 30 October 2019 (UTC)
- I agree, and I'm not sure why it was restored without gaining consensus here. That "Monsanto Papers" terminology is really only based in advocacy or WP:FRINGE related sources, so giving the term credence here is running into a few policy problems including WP:DUE. There's also a WP:COATRACKING issue going on in trying to fit in examples of the term being used.
- One is the sentence 'In 2017, internal documents from Monsanto, later called the "Monsanto Papers"'. The plays into into the due weight issues as we don't have sources to present that in Wikipedia's voice. If that had been really differentiated in mainstream use as opposed to mostly being in the advocacy side of things, there could be a case, but we don't have that. We also need to be especially mindful of legitimizing fringe viewpoints that try to get around the scientific criticism in the subject by trying to create controversy with terms like the "Monsanto Papers". When the IARC has been heavily criticized and contradicted by nearly every other independent scientific agency out there, we need to differentiate between advocacy source posturing versus legitimate problems Monsanto had. The other is the Le Monde article. The title isn't particularly important here, and it's just lumping together two loosely named things at best. There's also a bit too much exposition in that sentence instead of focus on the subject at hand, and some of the claims aren't quite matching up with the source either.
- At a minimum, the Monsanto Papers header should remain removed and the text put back under Glyphosate#International_Agency_for_Research_on_Cancer as it was before to keep the focus. How Monsanto responded and bits about that was supposed to be briefly handled under that main section, and I'm not sure a Monsanto response subheader is really needed. That said, that would be a much more NPOV header than the Monsanto Papers. Corporate responses are best left to a minimum to stick to what independent sources actually say instead though as this section would have a high likelihood of devolving into corporate responses and advocacy groups distracting from the core subject is we keep going down this route. Kingofaces43 (talk) 17:45, 31 October 2019 (UTC)
- Also with this source, it's a primary source with all the caveats about that even outside of MEDRS for this one. It hasn't really been cited, and the impact factor of the journal looks like it's pretty poor at > 1.[18][19] At least for a controversial subject like this, it doesn't look like it's something that would meet the due weight threshold for much if anything. Kingofaces43 (talk) 18:16, 31 October 2019 (UTC)
- I appreciate that Sj changed the header to something much more neutral. I think that solves most of my original concern. I wonder whether it's really WP:DUE to even discuss the term in that section, but I don't particularly mind and it's certainly no big deal to me. --Tryptofish (talk) 20:10, 3 November 2019 (UTC)
- 'Monsanto Papers' is now without doubt a mainstream term as per both scientific review papers: [1] and mainstream media around the world - a couple of example sources (many more available): "[2][3]BillyHatch2020 (talk) 00:36, 31 March 2020 (UTC)
I would also argue that your reviews on this page and on others related to glyphosate are not scientifically appropriate and are actually in the area of supporting corporate interests rather than the mainstream scientific opinion.BillyHatch2020 (talk) 00:36, 31 March 2020 (UTC)