14

This question and this question brought this to mind.

Questions like these, where the requester is asking a biased question not to gain actual knowledge but simply reinforce or spread their bias (note that I don't mean biased as "wrong", but just biased) are usually put on hold or closed, as with the two above, but when they accept a poor answer that only serves to reinforce their viewpoint, is there a way to remove acceptance of a bad answer without deleting it?

I understand this may become too subjective, but worth discussing.

  • I don't know if there is much that can be done about it, other than close bad questions. Moderators don't have the ability to remove the "accept" mark (SE employees do, but this happens only in cases of voting fraud and such), and even if they could, who decides what is "biased" or not? Not always so easy to determine. Personally, I think the entire "accept" feature is a misfeature, and will be happy when it will be entirely removed. – user11249 Mar 26 '18 at 16:37
  • @Carpetsmoker The accept won't be removed network wide, it provides a huge benefit on the more objective sites (like stackoverflow), but I'm not sure if it is a feature that can be changed on a site-by-site basis. I agree it is an interesting idea, I found this question on Meta.SE regarding Community Wikis only. – Jeff Lambert Mar 26 '18 at 16:55
  • 3
    it provides a huge benefit on the more objective sites (like stackoverflow) I don't agree with that at all either @JeffLambert. If you're asking a question then by definition you are not qualified to choose one answer as "the best". I have seen many Qs where the "accepted" answer is a bad one. There was another meta question that I can't find right now where Robert Cartaino (IIRC) explained that a number of SE employees are in favour of either removing the accept feature or not pinning answers on top, but that also other people were very much against it. – user11249 Mar 26 '18 at 17:00
  • Agreeing to disagree is one of the things I do best, my experience has been different. I have seen numerous questions as you state where the accepted one is bad, I just think there's just as many that are quite good and probably are indicative of which one helped the OP the most regardless of their (in)experience. – Jeff Lambert Mar 26 '18 at 17:08
  • 1
    Sure, but in those cases the accepted answers are almost always also the most upvoted ones @JeffLambert, so it seems to me that the "accept" mark is entirely redundant in those cases. – user11249 Mar 26 '18 at 17:14
  • 1
    @JeffLambert In my opinion, Politics.SE should be seen as an objective site too. There are plenty of subjective discussion forums about politics. Let's try to be the exception. – Philipp Mar 26 '18 at 17:21
  • 2
    @Carpetsmoker You seem to forget controversial topics. Many things are anathema for specific groups (e.g. Armenian genocide in Turkey), so if you scratch this topic and take stance, you will get downvotes independent of your position because either one or the other side is mad at you. So the "best" answer with votes will be the most agreeable and conciliatory one or the one which fits the view for most of the voters even if it is flat out wrong. While the most upvoted answer may be the best, beware the argumentum ad popolum fallacy. – Thorsten S. Mar 27 '18 at 12:55
  • @Philipp Agree. Other SEs are much stricter about fact-based questions and reference-based answers. I am working on a question that could go in law.SE, history.SE or politics.SE. The latter is my last choice. – Keith McClary Mar 31 '18 at 16:52
10

This problem is as old as Stack Exchange, I guess. In the end the question author has sole authority to decide which answer to accept. We can not override their decision, except by removing the answer altogether.

What you can do is:

  • Post a comment on the question where you express your concerns about their choice of accepted answer. Hope they reconsider. Assume good faith. New users might not yet realize that Politics.SE is not an opinion popularity contest.
  • If the answer has salvageable content between the bias, consider making an edit to remove the bias.
  • Downvote the answer to warn others that it might not actually be a good answer. Remember to leave a comment explaining what's wrong with the answer, so others know why it was downvoted and so the author can learn to write better answers.
  • Flag the answer as "very low quality" and hope the people on the review queue agree to delete it.
  • If the answer is so biased that it might offend people, flag it as "rude or offensive" and let the moderators decide if it should be deleted.

Also, if the question itself is biased, downvote the question itself and vote to close as "promotes or discredits a specific political cause". Closed and low-score questions disappear from the frontpage and from search results. Nobody will find the bad accepted answer if they don't even find the question.

  • 2
    You should probably mention the recent case when the OP has accepted a highly controversial answer (+26/-26 as of today) and indeed, the OP was gently informed about the problem and decided to retract their accept vote. – bytebuster Mar 27 '18 at 0:04
  • <Flag the answer as "very low quality" and hope the people on the review queue agree to delete it.> If a mod does not handle your flag first, like in my case. Also, your colleague disagrees "Please do not use flags in an attempt to reach consensus through the review queues" you should talk among yourselves – Federico Apr 6 '18 at 13:10
4

There's actually a gold badge for this: Populist, where your question gets way more upvotes than the accepted answer.

You're not going to get away from it, sadly. Accepted answers are, by definition, solely within the purview of the OP, and indicate the approval of the OP as answering their question, biases and all. An accepted answer does not necessarily indicate a good answer either.

4

I have a problem with this specific meta question because it seems to suggest that there are people who are not biased or viewpoints which are not biased. I don't think such people or viewpoints can exist for political questions.

I think what you really mean are people who are asking rhetorical questions because they already know the answer. Those people who are not really asking questions, but trying to spread their worldview. In this case the "promotes or discredits a specific political cause" flag is exactly the thing to ask for, but for all other cases I do not see any benefit for the following reasons.

First, the political landscape is vast and very different in different cultures, so what exactly do you mean with bias? A french anarchist, a German environmentalist, an US libertarian, a Zulu from the Inkatha party, an Israeli soldier, an Indian BJP party member, a Tunisian Muslim? Everyone will give you pretty good reasons for the worldview they have. Even the average Joe is pretty different for countries and cultures, so what should the benchmark be?

For all intents and purposes the best you can do to be impartial is citing only facts and documents. But even they have been selected from a pool of possible documents, they can be interpreted in different ways, (and yes, it is possible that several viewpoints can come to different valid(!) conclusions), you cannot read minds and you don't know the exact, often confidential, reasons for some decisions, so even with best effort you cannot make more than an educated guess.

So what if the "poor" answer is actually right or at least plausible, but violates your own bias? May I remind the threads where left-leaning people are complaining about right-leaning bias here and vice versa? If Politics.SE wants to be "unbiased" as best as possible, your question is, if taken word-for-word, a kind of oxymoron.

  • 1
    I had written an answer kinda along these lines but decided not to post it. I am glad you did. – Jeff Lambert Mar 27 '18 at 2:26
  • @JeffLambert Thanks, but I wonder about the reason? Fear of bad reaction? – Thorsten S. Mar 28 '18 at 4:17
  • I don't think I really captured the sentiment that underneath it all most questions (and all people in general) suffer from some form of bias, whether or not the asker or the community actually can pinpoint it or not. – Jeff Lambert Mar 28 '18 at 11:41
2

The accepted answer problem is very old and hard to tackle even for sites where "bias" is not that strong such as in Politics.SE (e.g. StackOverflow).

This meta question deals with why the accepted answer is on top by default, regardless of its quality. Cannot find one example now, but I saw an accepted answer with score -20 on SO simply because that answer became wrong after a while.

I think Philipp shows the correct actions to do which from my POV can be summarized to: the community should use already existing tools (downvote/flag/delete) to mark/remove bad content.

You must log in to answer this question.

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