2

StackExchange relies on community knowledge to arise at correct answers but this doesn't mean that the correct answer gets accepted by the OP. This is unavoidable due to the format used even though it violates the aim of providing correct data for a question.

On health.se, there are so few knowledgeable users that this compounds the problem so that the vote differences between the accepted incorrect answer, and the not accepted but correct answer might be nothing so anyone browsing the site might not get a clue.

I'd like to see a knowledgeable moderator step in and put a warning on these incorrect accepted answers. Health.se needs an exception due to the nature of the information being discussed.

And this is prompted by the obviously incorrect information in the accepted answer here by craig.feied

20
  • What do you mean by "a knowledgeable moderator"? Mar 5 '18 at 6:05
  • 2
    I"d say a high ranking user here ... but no one has even got to 10k :( So, a medical professional instead. And I accept the irony that the answer I'm complaining about is from an emergency room doc. Mar 5 '18 at 6:11
  • An ER doc stated that 2 g of ACAP was the daily limit? Wow. We'd be giving N-acetylcysteine left and right. Mar 5 '18 at 6:21
  • Exactly. That's why we need knowledgeable mods. Mar 5 '18 at 6:28
  • 1
    @GrahamChiu - One, this is completely against stack exchange model/policy. Two, as seen by your exchange with anon in the diabetes question, even experienced experts will differ in opinion. Would you as a mod have unaccepted the answer because you feel it is incorrect? Unfortunately, this is how it is across the network. You would need to raise the question on the SE meta if you feel strongly that it needs to change.
    – JohnP Mod
    Mar 5 '18 at 14:32
  • @JohnP - "...as seen by your exchange with anon in the diabetes question, even experienced experts will differ in opinion." Good point! Who is the expert? I agree with you. Meta SE is the place for this. Mar 5 '18 at 15:03
  • I wrote an answer to your question, then decided not to post. @JohnP answered in a comment. This is the SE Model. This is an SE site. This can't be addressed locally. Mar 5 '18 at 15:08
  • @JohnP - Didn't mean to sabotage your link, but since the erroneous comment was removed, my comments were no longer relevant. Mar 5 '18 at 15:12
  • 1
    @GrahamChiu - I guess I'm asking why you think this person is an ER doc. Really, an ER doc knows about ACAP/ASA overdoses. We know the magic number for healthy individuals and kids, and how to use the nomogram, etc. The answer is littered with misinformation. I don't think this person is an ER doc. Not everyone who claims to be something is being truthful. Also, there is a mod MD. She just happens to believe in the model. Mar 5 '18 at 15:23
  • Check his profile Mar 5 '18 at 16:44
  • Wow... wait where is the mod MD?
    – DoctorWhom
    Mar 5 '18 at 17:04
  • 1
    lol I actually meant to ask "who" it is, but "where" actually might fit better.
    – DoctorWhom
    Mar 5 '18 at 17:04
  • I think it's Susan but she doesn't have a profile. Mar 5 '18 at 17:30
  • @JohnP From where I am, it looks like the SE model generally works. But health.se hasn't reached a critical mass so it's malfunctioning. To ignore that just because it's against policy defies my common sense. Mar 5 '18 at 17:57
  • 1
    @GrahamChiu - Understandable, but that's the way it is. Expand the scope. It makes no sense for a longtime life partner with no legal documents to be unable to make medical decisions for their partner. But, it's policy.
    – JohnP Mod
    Mar 5 '18 at 18:03
4

This a suggestion I do not oppose.

There is really quite an amount of problems addressed in that question.

First of all, the wisdom of the crowd is not always reliable in the first place. We should know that. Sometimes the asker is a plain idiot. That happens. A lot? I have posts on other sites with the most extreme voting difference where I provided a very detailed answer, referenced my sources and even ended up with a negative score "too long", "do not like what you wrote" etc. Despite giving very plain and clearly the best answer. Bandwagon waltzed through town, this misalignment will probably never be corrected. OP apparently never read my final version of the answer and accepted a short but bogus answer. An effect that worked twice in a similar way, though hopefully not with a bogus answer, in my favour, to be fair.

Sometimes the answers are bad.

The accepted answer always relies on one person and her preferences alone. "Correctness" is to be established through voting, not alone, but primarily. There are badges for that, "outperforming an accepted answer".

What is "correct"? This is in most cases not so obvious as this question makes it look like. Final verdicts are an enemy of knowledge acquisition. Just like an accept from a never returning user on a wrong or at least inferior answer is a thorn in the side of a Q&A, such a requested marking might stifle progress of knowledge.

Who is knowledgable? In principle this is not reliably addressable here. I will not disclose any real information or even credentials that I have, neither should anyone. And if anyone does it, were are the guarantees that this information is correct and not just an impostering appeal to authority of unknowable validity? Even experts can err, and I have seen my share of this kind in all fields and walks of life.

Since voting is the correcting principle in this system of merits on SE, we are thrown back to not only our lack of real experts, medical professionals etc. We are thrown back on our sheer lack of core user base. And what is really compounding this small user base is the voting laziness within that small group.

The answer that prompted this discussion can and should be seen from two different perspectives of correctness.

  1. Does it confirm to our guidelines of format, reference providing etc?
  2. Is its content acceptable, correct or good?
  3. How do you, the voter, assess its up or downvote worthiness?

The OP (Graham) for this discussion only argues about 2 and 3, which are reasons not to be dismissed, but the answer already fails bullet point 1! That alone is reason to criticise the answer and if not improved: downvote it.

We did not do that! Apparently not effectively enough, to say the least.

It is precisely a design goal of any SE site that one question generates more than one answer. The theory is that one answer will be better than the other answer. The answer that prompted this is exactly one of the few cases where and when on our site this design goal that is necessary to graduate a site out of beta is fulfilled numerically (really >2.5 are needed?) and is showing the merits or demerits of the answers exactly like the theory envisions it.

The theory is a bit flawed, like I pointed out in the introduction. But here it seems to work out even though the voter sample is n=small.

Yes, we need more users with a healthy background, like medics, students, practitioners or theorists in a variety of fields. But we need more users overall, more active users, more questions, more answers, more votes. That does not sound greedy or overly ambitious. It is just a sign of severe deficiencies. The proposed marking for accepted but incorrect is then one supplement where we need much more healthy diet, that is life and style.

Asking for special treatment in this situation seems like a futile begging trip to the developers or staff at SE.

Your correcting comments are already in place and will hopefully stay there.

18
  • 2
    +1. Health.SE is a poor fit for the SE model, that's the real problem. If it were somehow limited to medical professionals of all kinds, maybe. But it's not, and most people who visit will still believe what they want to believe. As to fake appeals to authority, that shouldn't matter. It doesn't matter if I'm a doctor or not. My answers don't rely on that tactic; I used sources to back up my answers. Which is why answers should require reliable sources. Mar 5 '18 at 16:21
  • I think the SE model is a great idea, but it has its own fundamental flaws. That idea makes health a great fit. In theory. We do not give out advice or diagnosis; we are a science-oriented educational site. A very little Cochrane, sifting and gathering through the available evidence both good and bad, high powered or anecdotal etc. – Can the fundamental SE weaknesses be compensated or is this all our fault and misfortune (lack of site dynamic). Mar 5 '18 at 16:32
  • 3
    The problem is that science is not a popularity contest. The crowd is often wrong. And yes the answer failed point one which I pointed out to the answered. Mar 5 '18 at 16:48
  • "Can the fundamental SE weaknesses be compensated [for]..." No, they really can't, not for a "serious" Health site. When I was a mod, I used to check the sources on wonky answers, and usually the source simply didn't say what the OP was claiming it did. So even with sources, we would need enough knowledgeable users to recognize the wonky answers, comment and downvote. That hasn't happened yet, and will probably never happen if history repeats itself. Mar 5 '18 at 17:02
  • Another problem is that science should not be a pop-contest, but it all too often is. "Science progresses one funeral at a time." And sometimes it is quite rightly compared to a pop context. What else is "scientific consensus"? – I get your point, don't get me wrong on this. But the consensus model is what I mean with "SE-idea". That answers here have to be fact checked and peer reviewed just again points to "more users needed". Mar 5 '18 at 17:08
  • "...is this all our fault and misfortune (lack of site dynamic)." It is not our fault; it is the nature of the beast that Health attracts (very) poor questions and answers, and that no one likes to need to cite the literature. Some docs got uppity when asked for sources and left for that reason. The reasonable ones got bored with the penis/masturbation/can she be pregnant?/other stupid questions that have littered the site. Mar 5 '18 at 17:09
  • 1
    "That answers here have to be fact checked and peer reviewed just again points to "more users needed"." We interpret differently. I don't think we need more users. I think we need better, more interesting questions, and that isn't happening. Mar 5 '18 at 17:12
  • That uppity thing is a whole other story. That means these "docs" want to rely on (unprovable) authority and do not give out the evidence, the knowledge for what this site is built?. I enjoy being corrected by better evidence, logic and sources. – Very unfortunately I have too few counterarguments for your "questions problem", leading me to just agree on that one on its own. Mar 5 '18 at 17:16
  • I personally don't think every answer needs a reference. There are some things so obvious that they're unnecessary. Need to apply common sense here. Mar 5 '18 at 17:23
  • @GrahamChiu Of course. Only, that is possibly a twofold problem: 1. against site policy, if we manage to establish a stringent one, that is. 2. not every aspect needs a reference, but if the entire "answer does not need one" than it likely falls under "question is so basic, we should close it, not answer it" (aka "lack of prior research"). Common sense cannot and should not be excluded, but I think we were better of with largely spelled out rules which we and others can follow or apply. Mar 5 '18 at 17:31
  • The problem with references is where do you stop? Do you reference every single statement you make as if you were writing a paper? Mar 5 '18 at 17:33
  • 1
    @GrahamChiu No, certainly not. That's were common sense reigns supreme. I tried to write on this here and also here. You should add your views on those, too! Or point to other meta-posts which also address this. Mar 5 '18 at 17:41
  • 2
    @GrahamChiu - I disagree. It's common sense to an MD maybe, but not to everyone, and references allow the reader to learn more if they are interested, which is a valuable service. I agree with LangLangC here. Also, references are easy to find if one is familiar with the literature at all. Mar 5 '18 at 19:57
  • @anongoodnurse I totally agree. The SE model is a really bad fit, but also I believe that this site has little future. (that's why I'm in-coma). Really, we are lacking the critical user mass since I joined (>1 year ago), we are receiving way too many bad questions and there isn't a broad interest in this small failing meta site, so we don't get support from CM's who could maybe advise us with their experience.
    – Narusan
    Mar 6 '18 at 16:59
  • @Narusan-in-coma - That makes me smile. That answer was posted with no references, then the OP used HUFF POST as a reference. No, it's not the CMs' that help make a site. It's dedicated users. Unless there is popular support for a new mod, which I don't think there is. Mar 6 '18 at 20:20
2

Moderators are not arbiters of correctness, and if you start down this path you're going to be forever arguing corner cases. If an answer is completely, obviously wrong, then you don't need a label; if it's more nuanced, you place too much of a burden on the moderators to decide whether this one needs a warning but not that one. I'm a moderator on Worldbuilding, where we have one tag (hard-science) that has special requirements, clearly spelled out, and sometimes it's still hard to decide whether a particular answer is on the wrong side of the line and gets a post notice. Your definition of "wrong" will probably be fuzzier than ours.

Instead, work on the messaging about what the checkmark means, which is only that the person who asked the question liked that answer best. There are proposals on Meta Stack Exchange to always sort by votes (not pinning the accepted answer), to "age out" acceptances over time, and to remove the concept of acceptance entirely, probably among others. If you don't like the system (I'm personally in the "sort by votes, don't pin" camp), try to change it and in the mean time explain it clearly. But asking moderators to be judges of correctness will bring more heat than light.

2
  • 4
    +1. I agree that mods should not be responsible for the correctness of the answers; that was the biggest burden I gave myself as a mod, and I crumbled under the weight. I don't think our definition of wrong is fuzzier, though. Mar 6 '18 at 5:18
  • 2
    @anongoodnurse I just meant for that one tag, not our site in general. But it's a side point -- I'm certainly not trying to play the "which site is harder" game. :-) More seriously, I don't want to see good moderators crumple under the burden of having to be judges of correctness and dealing with the ensuing arguments. Mar 6 '18 at 14:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .