Is a bad answer better than no answer at all?

Question

An active user posted an answer to this question: How to increase my gut bacteria? Are there any drugs available? It has been deleted now, but a comment made by this active user gave me pause.

The user answered that eating yogurt was a good way to restore gut bacteria, and listed the National Yogurt Association's official website as the source. I commented that this source may make the answer a bit suspect, and asked for a better source to back up the answer.

Why do you believe the statement they make are factual? That might help. Any website with an economic interest should not be given the same consideration as reliable sources.

The user responded:

You want an honest answer? The honest answer is I didn't find the question compelling enough to invest the time in answering well, but it had the earmarks of one that would languish unanswered so I fired off an easy answer.

I have kept up on the scientific literature about gut microbiota because it interests me, and I know the answer is not cut and dry, but the comment took me by surprise.

Is a bad answer better than none at all? If so, why? If not, what can be done to discourage bad answers if even seasoned users don't follow site guidelines?

_{This view - that a bad answer is acceptable - in not isolated to this user.}

Answer 1

"Bad" is very general. In this case though, I think it's important to address the fundamental problem, which is laziness. Repeating what others have said, this site is health- it's a serious topic. It's not for hand-wavy logic and guesses or mumbled rumors.

We generally should not accept laziness, but as a beta community, we should still be patient with posters.

If some people get discouraged from questions and answers being difficult, then it's fine. Easy questions have google. People should welcome adversity to their answers on this site, as in true science: theories, studies, and information constantly challenged and improved. There are tons of things that we just don't have enough studies to prove, and that's simply something people need to accept.

If we're not sure, then we MUST have sources. We need to see legitimate proof of what is said. This is the part that troubles people. They want to post what seems to be an easy answer, but in fact, who knows how complex it could be? In health, there are often mysteries that we're continuing to unravel, and what seems to be may not be so. People need to realize that we just don't have all the answers, so the ones that do exist need to be substantiated with acceptable evidence.

If we think an answer is 100% correct... we STILL NEED SOURCES. "Sounds good" logic introduces bias. We should be wary of and try to fight against this consistently. Even in hospitals, (as much as you'd hate to hear this), doctors can and will make mistakes. Humans are fallible and require a system of accountability. E.g. - (http://qz.com/617173/childhood-obesity-is-linked-to-one-of-the-healthiest-foods-pregnant-women-can-eat/) which is about (http://archpedi.jamanetwork.com/article.aspx?articleid=2491661)

Look, even WITH sources, answers might still be wrong! But sources provide a basis on which to evaluate them.

Yes, yes, but who the hell has time for all of that? Our community should. Again, we are HEALTH.se. We need to hold ourselves to a strict standard if we genuinely want to help others and make a difference here.

Now, with all that being said, I mostly agree with januarium's suggested policies.

We're still in beta, so we need to be patient with all users. We should first leave comments, pointing out the problems with their answer, then come the down-votes if the OP fails to fix their post after some period of time. Finally comes flagging and deletion.

Answer 2

It depends on how bad the answer really is.

• If its link only or very, ridiculously small and not helpful then that may call for removal. The same thing for posts with bad advice. I would suggest a comment though first so that the user knows why and how to improve.

• If you consider a answer to be bad then I usually take it as my job as a concerned site member to triage it and find out whether it needs improvement or it can just go.

• If the post answers the questions and has some good research in it. Anything to be a saving grace, then it should be kept. A downvote and a comment explaining why and improvement should be dispensed.

• So yes, some bad answers can stay and are better than no answer at all. But only if they have potential. This site is trying to get sound advice, bad answers = bad site = less traffic = no site. So tough love has to be given and to even the solid users they need to be shown the ropes, as well. As a user that has just picked up more activity, I am learning the ropes as well.

In the Help Center I think it says that we make the site that we want. That said we have the option to crack down on them:

• Downvote their answer and explain why. All of the downvotes I give I usually give a comment with it explaining improvement. Once the improvement is done I remove the downvote and move on. On other sites I have been on I quote the help center and scope at the offender to help them and show them the error. Downvotes just make some people angry and cause them to leave the site. That is why I usually leave a comment and downvote later if they haven't gotten the message.

Answer 3

The main problem as I see it is that the rules by which we judge answers isn't going to work well. When considering a scientific paper for publication it is justified to demand that every nontrivial statement be backed up rigorously and therefore it is necessary to have a rigorous peer review system to evaluate scientific papers. However, on this site, there is no peer review system before publication, there is only going to be review after an answer is posted. Also we won't do any research as rigorously as when writing for a scientific journals. This means that there should be far more tolerance toward having fundamental discussions about the content of answers posted here.

Posting an answer here is analogous to having a well motivated argument in a scientific discussion, there likely will be some points in it that need to be ironed out before it becomes rigorous enough to include in a peer reviewed article. At least that's the case on all the other StackExchange sites except here, and that's why things are not working well here.