User talk:Colin/Archive 10
{{talk archive}}
MEDRS & tox
Hi Colin
I understand you are taking a wikibreak, but I have been looking for your input. I posted a too-long section in talk-MEDRS here about what I see as a widespread problem in Wikipedia. Namely, that basic researchers publish primary "tox" studies and draw big conclusions from them, that toxicologists would never draw, and that non-toxicologists publish reviews of those primary studies and also draw strong conclusions from them. Under current MEDRS, content based on those reviews is perfectly fine. However, reviews by toxicologists throw out many primary studies by basic researchers because the study design makes the results un-usable for tox (e.g. studies where a compound is injected instead of given orally may be useless when the main human exposure is oral). The result of the current MEDRS policy is that we get content in Wikipedia articles (especially on chemicals like various plastics, and herbicides, etc) that is scary and probably wrong -- exactly what MEDRS is meant to avoid. I think we should amend MEDRS to specify reliable sources for tox content. Something like (very draft-y here): "Sources that are most appropriate for content that summarizes results of in vitro or animal studies of the toxicity of products, are secondary and tertiary sources written by toxicologists. Primary studies should not be used to support content about toxicity." This may be a dead letter as whatamidoing seems to have a very different perspective than I do. However, I think this is important and may want to keep working to make it happen but would want more support. What are your thoughts on this issue generally, as well as the idea of amending MEDRS to deal with it? thx Jytdog (talk) 04:05, 30 April 2013 (UTC)
:See Dreyfus model of skill acquisition (the article is a mess but the ideas helpful). Our problem often is that editors come to WP with the whole range of skill and sometimes with conflicting skills (such as academic writing). Simple advice such as "avoid primary research papers as sources" will be helpful to many beginners and appropriate for a huge amount of WP writing. Sometimes, discussions on MEDRS can fall into the trap where experience at Wikipedia combines with a desire to be completely accurate in case someone pick holes in one's argument. The result can be "correct" but unhelpful advice or commentary. MEDRS is trying to be useful for a wide range of medical articles and for Wikipedians with a wide range of abilities and access to sources. Sometimes people look for a binary answer to a complex problem. Like whether a source is primary or secondary (it depends). Or whether some source is "MEDRS compliant" and presumably then we can include whatever it says into the article. Or if some issue is governed by MEDRS at all. It can be helpful to step back from the specifics of the question and look for general advice in MEDRS or other policy. I find WP:WEIGHT wins many arguments. Certainly MEDRS could be improved -- it is too long for a start.
:Wrt toxicology, I know nothing about the subject. Are the deficiencies you mention documented by commentators on the field perhaps? For example, some of the problems of fMRI research or evolutionary psychology or drug studies are well documented. Perhaps there is some reason we should single-out toxicology for mention? However, MEDRS already says "written by experts in the relevant field and from a respected publisher". So doesn't that already place issues of toxicity written by toxicologists as more likely to be valuable sources than the writings of non-toxicologists? But where these issues require abilities in two or more domains, that does become tricky. I'm not sure there is a general solution.
:Can your problems with specific sources be better handled by stepping back and asking, per WP:WEIGHT, if I collect some of the best reliable sources on this subject, what do they say about it and in what proportion? Often it seems folk start with a source that says something they want to say, and then argue about the validity of that source in isolation. Rather than trying to remove the claim because the source is bad, perhaps the claim doesn't deserve mention per WP:WEIGHT or its inclusion should be minimal. Colin°Talk 12:20, 30 April 2013 (UTC)
::Thanks for your quick reply! I had the same idea, that I could bolster my argument with a review article or tox textbook, discussing the problem broadly. I am still looking for that. I hear you that the text you quote from MEDRS addresses it, and I agree, but you may have seen that whatamidoing believes that any medical professional can make reliable statements on tox. I don't understand what you say about something requiring "two or more domains" -- within the field of toxicology there are people who specialize in neurotoxicity, cardiotoxicity, etc. From my perspective having the toxicology framework first is essential for this stuff. I hear you too, on using the WEIGHT argument but especially dealing with editors with a a strong environmentalist bent, those arguments are very hard to make, since in the world they see (weltbild rather than weltanschauung) and the sources they read, these toxicology issues loom very large and should have more weight than anything. I will continue searching for the kind of general discussion by toxicologists on basic research studies that you mentioned - will let you know what I find! There is no hurry on this. Thanks again. Jytdog (talk) 16:36, 30 April 2013 (UTC)
:::The "two or more domains" comment refers to issues that require experts in multiple disciplines. Often people writing a comprehensive article need to cover areas they are particularly expert in and areas they are weaker. Someone writing outside of their speciality can write something quite reasonable. Unless they are in a controversial area, these experts-in-something are probably more likely to be useful than the experts-in-nothing that write our newspapers. There are rogues, of course. Nobel prize winners who use their fame to write nonsense but also subject-experts who are just plain maverick. I'm not really sure what whatamidoing is getting at wrt neurologists vs toxicologists. Surely both terms are too vague to be useful in this argument. Some toxicologists will know nothing about nervous diseases and plenty neurologists will remember little about how chemicals affect the nerves (they spend their day seeing and treating patients, each of which contains billions of neurons). I don't know much about toxicology, but if it provides a scientific framework for these sort of topics then that is surely to be valued. There are certainly many pitfalls and poor science is published -- something that our best review authors know to discard or comment-on when reviewing the field.
:::You need to bear in mind that the imperfect world cannot be fixed on Wikipedia. We are at the mercy of the body of literature published and you won't win the battle on every article. If experts generally disagree on something, we will need to reflect that -- we can't just pick the expert we think is best and take their views alone. Colin°Talk 19:31, 30 April 2013 (UTC)
::::Thanks for this response. I guess it comes down, a bit, to what "type" of study. As per MEDRS (as you know, I am telling you that I know) "'Assessing evidence quality' means that editors should determine the quality of the type of study." I think this might be a useful place to add a toxicology review as a type of study -- because the "best" type of study, a "meta-analyses of randomized controlled trials (RCTs)" - will never be available with respect to tox. It doesn't seem that MEDRS wasn built taking tox into account and this is where it would be great to have some structure... btw I have gotten a couple of tox textbooks from the library and am reading, and plan to call a few today to ask about this question of "tox" studies done by basic researchers. Jytdog (talk) 15:08, 3 May 2013 (UTC)
POTD notification
File:Electric steam iron.jpg]]
Hi Colin,
Just to let you know that the Featured Picture :File:Electric steam iron.jpg is due to make an appearance as Picture of the Day on June 16, 2013. If you get a chance, you can check and improve the caption at Template:POTD/2013-06-16. Thank you for all of your contributions! — Crisco 1492 (talk) 23:02, 30 May 2013 (UTC)
:Hi Colin,
:This is a very cool picture. Thanks for sharing it. WhatamIdoing (talk) 03:50, 14 July 2013 (UTC)
{{-}}
The Wikipedia Library now offering accounts from Cochrane Collaboration (sign up!)
{{WP:TWL/Header}}
Cochrane Collaboration is an independent medical nonprofit organization consisting of over 28,000 volunteers in more than 100 countries. The collaboration was formed to organize medical scholarship in a systematic way in the interests of evidence-based research: the group conducts systematic reviews of randomized controlled trials of health-care interventions, which it then publishes in the Cochrane Library.
Cochrane has generously agreed to give free, full-access accounts to 100 medical editors. Individual access would otherwise cost between $300 and $800 per account. Thank you Cochrane!
If you are stil active as a medical editor, come and sign up :)
